1 00:00:15,276 --> 00:00:26,156 Speaker 1: Pushkin. For roughly five thousand years, people call themselves doctors 2 00:00:26,276 --> 00:00:28,436 Speaker 1: and pretended to know all sorts of things that they 3 00:00:28,476 --> 00:00:30,436 Speaker 1: didn't know, and were as likely to kill you as 4 00:00:30,436 --> 00:00:35,836 Speaker 1: to cure you. These doctors existed because sick people desperately 5 00:00:35,876 --> 00:00:39,636 Speaker 1: wanted to believe in them. Coaching feels the same way 6 00:00:39,636 --> 00:00:44,796 Speaker 1: to me. For decades, people just sort of hoped that 7 00:00:44,836 --> 00:00:47,036 Speaker 1: if a man was hollering at them, he must be 8 00:00:47,076 --> 00:00:51,236 Speaker 1: helping them to win. Maybe he was sometimes, but that's 9 00:00:51,236 --> 00:00:53,796 Speaker 1: not my point. My point is that even if coaches 10 00:00:53,836 --> 00:00:56,756 Speaker 1: have no effect on performance, even if they're doing more 11 00:00:56,796 --> 00:01:00,836 Speaker 1: harm than good, we might still insist on having them 12 00:01:00,876 --> 00:01:03,676 Speaker 1: because we need someone on our side to believe in. 13 00:01:07,476 --> 00:01:10,916 Speaker 1: But coaching's changing the same way medicine changed one hundred 14 00:01:10,996 --> 00:01:17,996 Speaker 1: years ago. Coaches are discovering science, and science is discovering them. 15 00:01:18,076 --> 00:01:22,196 Speaker 1: I'm Michael Lewis and This is Against the Rules, a 16 00:01:22,276 --> 00:01:26,596 Speaker 1: show about various authority figures in American life. This season 17 00:01:26,676 --> 00:01:29,676 Speaker 1: is about the rise of coaches, and this episode is 18 00:01:29,716 --> 00:01:45,636 Speaker 1: about data and pitching. A while back in two thousand 19 00:01:45,636 --> 00:01:48,636 Speaker 1: and three, I published a book called Moneyball. It's about 20 00:01:48,676 --> 00:01:51,716 Speaker 1: how the Oakland A's baseball team had used data analysis 21 00:01:51,716 --> 00:01:54,276 Speaker 1: to get an edge on everyone else. They were a 22 00:01:54,356 --> 00:01:57,476 Speaker 1: poorly funded team in a small market. They had no 23 00:01:57,556 --> 00:02:00,276 Speaker 1: money to spend on players, but their new and better 24 00:02:00,316 --> 00:02:05,076 Speaker 1: statistics enable them to value baseball players more accurately, so 25 00:02:05,116 --> 00:02:07,436 Speaker 1: they could sell the players that were overvalued and trade 26 00:02:07,476 --> 00:02:10,036 Speaker 1: for the ones that were undervalued. I remember at the 27 00:02:10,076 --> 00:02:12,916 Speaker 1: time being shocked at the notion that baseball players could 28 00:02:12,916 --> 00:02:15,916 Speaker 1: be misvalued. I mean, baseball players have been doing the 29 00:02:15,956 --> 00:02:18,476 Speaker 1: same job for a century, out in the open, in 30 00:02:18,516 --> 00:02:27,876 Speaker 1: front of millions of people. But suddenly, all over a 31 00:02:27,996 --> 00:02:30,356 Speaker 1: thing that had been done a certain way forever was 32 00:02:30,436 --> 00:02:35,876 Speaker 1: now being done a totally different way. Everyone in baseball 33 00:02:35,916 --> 00:02:38,676 Speaker 1: started using data and getting all sorts of insights from it, 34 00:02:39,356 --> 00:02:42,076 Speaker 1: and the insights led not just to better valuations of 35 00:02:42,116 --> 00:02:45,876 Speaker 1: baseball players, it eventually led to a new kind of coaching. 36 00:02:46,396 --> 00:02:49,236 Speaker 1: In the past, it used to be that many coaching 37 00:02:49,276 --> 00:02:52,836 Speaker 1: positions were almost a sine cure. That's Ben Lindberg, co 38 00:02:52,956 --> 00:02:56,396 Speaker 1: author of a book called The MVP Machine. It's about 39 00:02:56,436 --> 00:02:59,916 Speaker 1: a revolution in how the world's best baseball players get coached. 40 00:03:00,356 --> 00:03:03,596 Speaker 1: It was the coaches who were the manager's palace, his 41 00:03:03,756 --> 00:03:07,916 Speaker 1: drinking buddies would become the coaches, and there wasn't that 42 00:03:08,116 --> 00:03:11,236 Speaker 1: much coaching going on at the major league level. It 43 00:03:11,276 --> 00:03:14,436 Speaker 1: was sort of reinforcing lessons that had already been taught 44 00:03:14,836 --> 00:03:17,996 Speaker 1: and keeping guys in line, but there wasn't that much 45 00:03:18,036 --> 00:03:21,556 Speaker 1: expectation that coaches would improve players once they got to 46 00:03:21,636 --> 00:03:24,836 Speaker 1: that level. But that was about to change big time. 47 00:03:25,956 --> 00:03:28,436 Speaker 1: Ben was part of a huge and growing crowd of 48 00:03:28,556 --> 00:03:31,796 Speaker 1: data geeks outside of baseball who spent lots and lots 49 00:03:31,796 --> 00:03:34,916 Speaker 1: of time analyzing players and building models to try to 50 00:03:34,956 --> 00:03:39,836 Speaker 1: predict their performance. As sophisticated as the statistical projection models are, 51 00:03:40,156 --> 00:03:42,876 Speaker 1: they'll only really look at the player's past performance and 52 00:03:43,036 --> 00:03:45,956 Speaker 1: his age and maybe some comparable players from the past, 53 00:03:45,996 --> 00:03:48,756 Speaker 1: and they'll spit out a projection that say, well, he 54 00:03:48,836 --> 00:03:51,196 Speaker 1: was this good in the past few years, and we'll 55 00:03:51,196 --> 00:03:53,316 Speaker 1: adjust for the ballpark, and here's how old he is, 56 00:03:53,356 --> 00:03:56,356 Speaker 1: and so here's our median projection for him, and so 57 00:03:56,396 --> 00:03:58,356 Speaker 1: all of a sudden, there players who are just busting 58 00:03:58,356 --> 00:04:02,076 Speaker 1: out of those projections exactly, And a projection system would 59 00:04:02,156 --> 00:04:04,716 Speaker 1: never forecast that. It might say that someone's going to 60 00:04:04,756 --> 00:04:07,516 Speaker 1: get a bit better, a bit worse, but typically it 61 00:04:07,556 --> 00:04:09,756 Speaker 1: won't say that someone is going to do something that's 62 00:04:09,756 --> 00:04:13,196 Speaker 1: completely out of line with their past performance. In other words, 63 00:04:13,196 --> 00:04:15,636 Speaker 1: teams had gotten really good, or at least a lot 64 00:04:15,676 --> 00:04:19,316 Speaker 1: better at evaluating the potential of all their players, but 65 00:04:19,396 --> 00:04:23,236 Speaker 1: the players were still playing better than expected, so much 66 00:04:23,276 --> 00:04:26,236 Speaker 1: better that the analysts were a bit suspicious. But the 67 00:04:26,276 --> 00:04:30,036 Speaker 1: last time you saw this was with the steroids era, 68 00:04:30,396 --> 00:04:32,676 Speaker 1: that all of a sudden, people players were performing in 69 00:04:32,716 --> 00:04:36,276 Speaker 1: ways that the projection models would never have guessed. And 70 00:04:36,356 --> 00:04:39,396 Speaker 1: you're so you're kind of seeing it, but without an 71 00:04:39,916 --> 00:04:44,756 Speaker 1: explanation as obvious as steroids exactly. Yeah, then started looking 72 00:04:44,756 --> 00:04:47,356 Speaker 1: into what these players were doing. The ones who were 73 00:04:47,436 --> 00:04:53,116 Speaker 1: dramatically exceeding the analysts expectations, the overperformers, all had something 74 00:04:53,116 --> 00:04:57,196 Speaker 1: in common. Coaches who use new technology. One of the 75 00:04:57,316 --> 00:05:01,756 Speaker 1: really big innovations has been the high speed camera. So 76 00:05:01,916 --> 00:05:06,996 Speaker 1: a company called Edgrotronic, which developed these cameras for scientific purposes, 77 00:05:07,316 --> 00:05:09,876 Speaker 1: has found that much of its business has come from 78 00:05:09,916 --> 00:05:13,116 Speaker 1: baseball teams because baseball teams have found that if you 79 00:05:13,196 --> 00:05:16,516 Speaker 1: train these high speed cameras on players, you can perceive 80 00:05:16,636 --> 00:05:20,276 Speaker 1: things about players movement that they didn't know about themselves. 81 00:05:20,756 --> 00:05:23,836 Speaker 1: And the coaches using these cameras were very different from 82 00:05:23,836 --> 00:05:28,756 Speaker 1: the old school baseball coaches, the Sinecure guys. For a start, 83 00:05:28,796 --> 00:05:32,836 Speaker 1: the new coaches weren't former big league players. In some cases, 84 00:05:32,996 --> 00:05:35,436 Speaker 1: they didn't even know any big league players. My name 85 00:05:35,476 --> 00:05:38,196 Speaker 1: is Kyle Bodie. Like this guy, I was twenty two 86 00:05:38,276 --> 00:05:42,356 Speaker 1: years old started coaching little league and I realized quickly 87 00:05:42,556 --> 00:05:45,716 Speaker 1: that I just didn't know anything about coaching. Kyle had 88 00:05:45,756 --> 00:05:48,556 Speaker 1: just moved from Ohio to Seattle, where he landed a 89 00:05:48,596 --> 00:05:51,636 Speaker 1: part time job as a little league coach. Yep, that's 90 00:05:51,636 --> 00:05:54,556 Speaker 1: how he started in life as a little league baseball coach. 91 00:05:54,876 --> 00:05:57,556 Speaker 1: Ed played some my father was a coach, but I 92 00:05:57,676 --> 00:06:00,076 Speaker 1: figured I owed it to the kids to learn just 93 00:06:00,116 --> 00:06:02,596 Speaker 1: a little bit more about keeping their arms healthy. And 94 00:06:02,676 --> 00:06:05,956 Speaker 1: he had some questions like how many pitches should a 95 00:06:06,036 --> 00:06:08,356 Speaker 1: kid be allowed to throw? And what was the best 96 00:06:08,356 --> 00:06:10,836 Speaker 1: way to throw him? I mean, I was once a pitcher, 97 00:06:10,916 --> 00:06:12,716 Speaker 1: and I spent half my life with my arm in 98 00:06:12,756 --> 00:06:15,356 Speaker 1: an ice bucket. To this day, I can't sleep on 99 00:06:15,396 --> 00:06:18,796 Speaker 1: my right shoulder throwing a ball overhand. It might look 100 00:06:18,836 --> 00:06:21,356 Speaker 1: like a natural and healthy thing to do, but it's not. 101 00:06:22,596 --> 00:06:25,276 Speaker 1: Kyle body looked around for research on the subject, but 102 00:06:25,396 --> 00:06:29,356 Speaker 1: unfortunately it was all very nonspecific kind of very academic research, 103 00:06:30,076 --> 00:06:33,276 Speaker 1: and then the training programs and the coaching programs that 104 00:06:33,316 --> 00:06:37,356 Speaker 1: were out there were very bland, not based on any 105 00:06:37,356 --> 00:06:40,916 Speaker 1: sort of evidence. It really shocked me, So Kyle started 106 00:06:40,956 --> 00:06:43,636 Speaker 1: to do research on his own. Then he got a 107 00:06:43,676 --> 00:06:46,196 Speaker 1: promotion from Little league to the freshman team at a 108 00:06:46,236 --> 00:06:50,076 Speaker 1: Seattle high school. But he found himself at war with 109 00:06:50,196 --> 00:06:53,716 Speaker 1: the JV and varsity coaches at the high school. They 110 00:06:53,716 --> 00:06:56,956 Speaker 1: were coaching the old fashioned way, telling players what to do, 111 00:06:57,836 --> 00:07:00,356 Speaker 1: hollering at them when they screwed up, praising them when 112 00:07:00,396 --> 00:07:04,316 Speaker 1: they didn't. Actively coaching athletes just typically makes them worse. 113 00:07:04,916 --> 00:07:07,876 Speaker 1: Just intervention is typically one of the worst things you 114 00:07:07,916 --> 00:07:10,596 Speaker 1: can do. What are the points of friction with the 115 00:07:10,676 --> 00:07:13,636 Speaker 1: old coaching model, Like, what specifically kind of things would 116 00:07:13,676 --> 00:07:17,716 Speaker 1: you do that were heretical. So informing the athlete that, like, 117 00:07:18,076 --> 00:07:21,036 Speaker 1: whatever they're doing is not good enough and then just 118 00:07:21,116 --> 00:07:24,116 Speaker 1: seeing how they change over time and how they will 119 00:07:24,156 --> 00:07:28,676 Speaker 1: self organize was a real heretical idea, right because most 120 00:07:28,676 --> 00:07:31,596 Speaker 1: coaches think that they have a lot to give to 121 00:07:31,676 --> 00:07:34,476 Speaker 1: the athlete, and my view on it still is today 122 00:07:35,116 --> 00:07:36,956 Speaker 1: is that they're good enough, like we just need to 123 00:07:36,956 --> 00:07:38,876 Speaker 1: give them the right direction and let them figure it 124 00:07:38,916 --> 00:07:40,676 Speaker 1: out for the most part. You ever read a book 125 00:07:40,716 --> 00:07:43,036 Speaker 1: called The Inner Game of Tennis? I have, Yeah, it's 126 00:07:43,036 --> 00:07:48,956 Speaker 1: one of my favorites. Anyway, In his first year coaching, 127 00:07:49,076 --> 00:07:51,916 Speaker 1: Kyle's freshman team won as many games as they lost, 128 00:07:52,316 --> 00:07:55,556 Speaker 1: which was actually pretty great. That season, the school's varsity 129 00:07:55,556 --> 00:07:58,676 Speaker 1: and JV teams were losing most of their games. But 130 00:07:58,716 --> 00:08:00,796 Speaker 1: at the end of the year, the head coach fired 131 00:08:00,916 --> 00:08:04,596 Speaker 1: Kyle because the other coaches hated his methods. They thought 132 00:08:04,596 --> 00:08:07,956 Speaker 1: he had no clue how to coach. That didn't stop Kyle. 133 00:08:08,236 --> 00:08:12,316 Speaker 1: He doubled down on his approach, and he wound up 134 00:08:12,356 --> 00:08:17,516 Speaker 1: building what amounted to a baseball bionic manfactory drive line Baseball, 135 00:08:17,556 --> 00:08:22,436 Speaker 1: he called it. So describe to me this laboratory you build. 136 00:08:22,836 --> 00:08:27,036 Speaker 1: In its current incarnation, it's about fifteen high speed motion 137 00:08:27,076 --> 00:08:30,916 Speaker 1: tracking cameras. There's force plates in there to measure ground 138 00:08:30,956 --> 00:08:34,796 Speaker 1: reaction forces. Attracts every movement two hundred forty times per second, 139 00:08:35,356 --> 00:08:38,236 Speaker 1: and it's submillimeter accurate, so every movement can be tracked 140 00:08:38,236 --> 00:08:41,116 Speaker 1: down to at least one millimeter of accuracy, and usually 141 00:08:41,196 --> 00:08:48,276 Speaker 1: much better. Let's go one hundred point four ninety three 142 00:08:48,316 --> 00:08:52,476 Speaker 1: point seven the high speed cameras allow Kyle to measure 143 00:08:52,516 --> 00:08:56,236 Speaker 1: the speed of a pitcher's arm, among other things, the 144 00:08:56,356 --> 00:08:58,876 Speaker 1: faster a pitcher's arm, the faster the ball comes out 145 00:08:58,876 --> 00:09:04,676 Speaker 1: of it in theory one hundred three point zero. In practice, 146 00:09:04,796 --> 00:09:07,556 Speaker 1: not all pictures are able to translate their arm speed 147 00:09:07,676 --> 00:09:11,156 Speaker 1: into ball speed. Two guys with the exact same arm 148 00:09:11,236 --> 00:09:14,996 Speaker 1: speeds might throw very different fastballs. If someone's arm speed 149 00:09:15,076 --> 00:09:18,516 Speaker 1: is extremely high and the ball comes out at like 150 00:09:18,556 --> 00:09:22,316 Speaker 1: a lower predictive velocity based on like a regression algorithm, 151 00:09:22,836 --> 00:09:24,916 Speaker 1: that we know that there's some inefficiencies there that we 152 00:09:24,956 --> 00:09:27,316 Speaker 1: should be able to easily clean up. That is, you 153 00:09:27,316 --> 00:09:30,156 Speaker 1: could identify people with the god given talent to throw 154 00:09:30,196 --> 00:09:33,596 Speaker 1: a baseball because you had new insight into where that 155 00:09:33,676 --> 00:09:38,036 Speaker 1: talent came from. So when that happens, are you thinking, 156 00:09:38,236 --> 00:09:40,996 Speaker 1: there's this pool of players out there who have high 157 00:09:41,116 --> 00:09:44,876 Speaker 1: arm speed, low velocity that we can just fix. That's 158 00:09:44,876 --> 00:09:47,596 Speaker 1: all that I think about pretty much every day. And 159 00:09:47,716 --> 00:09:50,356 Speaker 1: could you like drive around with a little truck and 160 00:09:50,396 --> 00:09:52,356 Speaker 1: put people in the back of your truck and test 161 00:09:52,356 --> 00:09:54,676 Speaker 1: their arm speed? Is that what you would do? Yeah, 162 00:09:54,716 --> 00:09:57,076 Speaker 1: that's actually funny. We almost butt an RV to do 163 00:09:57,116 --> 00:09:59,916 Speaker 1: exactly that to drive around the country with our lab. 164 00:10:00,356 --> 00:10:02,676 Speaker 1: It turned out he didn't need to drive around looking 165 00:10:02,676 --> 00:10:08,116 Speaker 1: for people with this weird arm talent. They found him. 166 00:10:08,636 --> 00:10:10,756 Speaker 1: I real that I needed to make a change going 167 00:10:10,796 --> 00:10:13,076 Speaker 1: into the season. I was helping out my dad's team 168 00:10:13,076 --> 00:10:15,516 Speaker 1: and trying to make money in every way. That's Matt 169 00:10:15,556 --> 00:10:19,156 Speaker 1: Boyd Back in twenty seventeen. He was an unknown minor 170 00:10:19,236 --> 00:10:22,036 Speaker 1: league pitcher coming off a terrible season. He had a 171 00:10:22,036 --> 00:10:25,316 Speaker 1: below average fastball around eighty nine miles an hour. Oh, 172 00:10:25,436 --> 00:10:28,036 Speaker 1: and he had an arm injury. The way he was going, 173 00:10:28,076 --> 00:10:29,916 Speaker 1: he's about to be spending a lot more time with 174 00:10:29,956 --> 00:10:32,596 Speaker 1: his mom and dad. Over to the Christmas break, one 175 00:10:32,596 --> 00:10:36,516 Speaker 1: of my dad's players came back and he was ninety 176 00:10:36,516 --> 00:10:39,516 Speaker 1: two guy in high school and he was at Oregon State, 177 00:10:39,556 --> 00:10:42,316 Speaker 1: and after the fall he was ninety eight miles per hour. Oh, 178 00:10:42,396 --> 00:10:45,396 Speaker 1: I went, whoa, whoa, whoa, I go, what were you 179 00:10:45,396 --> 00:10:48,956 Speaker 1: doing down there? He's like, I did the drive Line program. 180 00:10:49,196 --> 00:10:52,516 Speaker 1: Drive Line. That's Kyle Bodie's lab. So at that point, 181 00:10:52,596 --> 00:10:54,796 Speaker 1: Kyle would have been kind of a local secret. So 182 00:10:54,836 --> 00:10:57,516 Speaker 1: someone all of a sudden has this kind of miraculous 183 00:10:57,996 --> 00:11:00,396 Speaker 1: jump in velocity. You hear about it, and you go see, 184 00:11:00,516 --> 00:11:02,356 Speaker 1: you call Kyle had had you ever heard of him 185 00:11:02,356 --> 00:11:05,716 Speaker 1: at that point? No? I hadn't heard of him. Matt 186 00:11:05,716 --> 00:11:10,636 Speaker 1: goes to seek Kyle in his Bionic manfactory and here 187 00:11:10,716 --> 00:11:14,716 Speaker 1: we have cameras, great our guns. I go up a 188 00:11:14,756 --> 00:11:17,596 Speaker 1: little stairwell and here's you know, a little eight foot 189 00:11:17,636 --> 00:11:21,916 Speaker 1: wide by ten foot high probably pitching lane tunnel created 190 00:11:21,956 --> 00:11:24,236 Speaker 1: with all this technology in there. And there's Kyle behind 191 00:11:24,236 --> 00:11:27,076 Speaker 1: a computer and he runs me through the program. At 192 00:11:27,116 --> 00:11:31,236 Speaker 1: that point, had you ever seen the technology that was there? No? 193 00:11:31,396 --> 00:11:33,156 Speaker 1: And I honestly I didn't. I couldn't even told you 194 00:11:33,316 --> 00:11:35,276 Speaker 1: right today what I saw in there. I couldn't even 195 00:11:35,276 --> 00:11:37,156 Speaker 1: told you what was going on a bunch of fucket 196 00:11:37,156 --> 00:11:39,796 Speaker 1: lids for some for balldrills. It looks like this, a 197 00:11:39,836 --> 00:11:42,756 Speaker 1: lot of gizmos, a lot of gizmos, a lot of wires, 198 00:11:42,796 --> 00:11:45,076 Speaker 1: a lot of interesting looking baseballs, a lot of lines 199 00:11:45,116 --> 00:11:48,316 Speaker 1: on a pitching mound and stuff. And you know, Kyle 200 00:11:48,356 --> 00:11:50,876 Speaker 1: explained to me what the concept of what he does, 201 00:11:51,436 --> 00:11:54,676 Speaker 1: and we talked about it, and then we just started 202 00:11:54,676 --> 00:12:01,716 Speaker 1: the program. Kyle put Matt Boy through a bunch of tests. 203 00:12:02,236 --> 00:12:04,836 Speaker 1: The big one was to test his arm speed, but 204 00:12:04,916 --> 00:12:08,996 Speaker 1: he never used the phrase arm speed. Kyle actually never 205 00:12:09,196 --> 00:12:12,036 Speaker 1: told Matt what he was testing for. He in our 206 00:12:12,116 --> 00:12:15,756 Speaker 1: lab tested higher than pretty much everyone and still almost 207 00:12:15,796 --> 00:12:18,756 Speaker 1: everyone to this day. He's just an excellent athlete. And 208 00:12:18,956 --> 00:12:21,516 Speaker 1: yet the ball velocity wasn't where it needed to be 209 00:12:21,756 --> 00:12:23,796 Speaker 1: or where it predicted it should be. So what was 210 00:12:23,836 --> 00:12:26,076 Speaker 1: the inefficiency? Like? What was what was he doing that 211 00:12:26,276 --> 00:12:28,196 Speaker 1: caused the ball to come out more slowly than what 212 00:12:28,276 --> 00:12:30,356 Speaker 1: you would have predicted? Right, it's really hard. We don't 213 00:12:30,396 --> 00:12:32,716 Speaker 1: know the rude cause yet. That's that's the actual interesting 214 00:12:32,756 --> 00:12:37,476 Speaker 1: thing is we're still studying, like why this happens. At 215 00:12:37,476 --> 00:12:40,756 Speaker 1: the lab, Kyle hands Matt these really heavy balls to throw. 216 00:12:41,636 --> 00:12:44,676 Speaker 1: He's found that when people throw a heavy ball, their 217 00:12:44,716 --> 00:12:47,596 Speaker 1: body naturally finds the most efficient way to do it 218 00:12:48,116 --> 00:12:51,236 Speaker 1: because it's so painful and uncomfortable to throw it inefficiently. 219 00:12:52,996 --> 00:12:57,676 Speaker 1: Matt basically moves into the Bionic manfactory and throws heavy 220 00:12:57,676 --> 00:13:01,516 Speaker 1: balls the entire off season. Then rejoins his minor league 221 00:13:01,556 --> 00:13:05,476 Speaker 1: team and you know, I get down there, I tell 222 00:13:05,556 --> 00:13:08,756 Speaker 1: him I'm I'm full progression off the mount and they're like, okay, 223 00:13:08,756 --> 00:13:11,956 Speaker 1: well let's see it. My first bullpen I'm I'm ninety 224 00:13:11,996 --> 00:13:15,676 Speaker 1: two to ninety five. Oh, and I think everyone's going, 225 00:13:16,476 --> 00:13:19,996 Speaker 1: what the heck happened? You know, and even I am, 226 00:13:20,036 --> 00:13:22,476 Speaker 1: I'm going, man, I'm throwing the baseball up in the zone. Now, 227 00:13:22,636 --> 00:13:24,596 Speaker 1: this is amazing. This is so cool. Like when all 228 00:13:24,636 --> 00:13:26,716 Speaker 1: of a sudden, you've got this new weapon. Yeah, when 229 00:13:26,756 --> 00:13:28,516 Speaker 1: you got this fastball that's coming out of your hand 230 00:13:28,516 --> 00:13:30,636 Speaker 1: three or four miles an hour faster than it usually does. 231 00:13:31,036 --> 00:13:33,956 Speaker 1: What do you notice in what happened and how hitters 232 00:13:33,996 --> 00:13:37,636 Speaker 1: respond and how the effectiveness hitters have against you. I 233 00:13:37,676 --> 00:13:40,956 Speaker 1: remember going in to Double A that season and I did. 234 00:13:40,956 --> 00:13:42,796 Speaker 1: We have a catcher named Jack Murphy who was about 235 00:13:42,836 --> 00:13:45,116 Speaker 1: four or five years older than me, and he went 236 00:13:45,196 --> 00:13:47,476 Speaker 1: up and told me, he goes, Maddie, you have a 237 00:13:47,476 --> 00:13:50,036 Speaker 1: new fastball. You need to pitch off it. I remember, 238 00:13:50,076 --> 00:13:52,276 Speaker 1: I was kind of scared. I was like, what, I've 239 00:13:52,276 --> 00:13:54,436 Speaker 1: never done that in my life. I'm always fastball, change up. 240 00:13:54,436 --> 00:13:56,116 Speaker 1: And then I mixed my curveball, and you know, but 241 00:13:56,676 --> 00:13:58,356 Speaker 1: he challenged me, and all of a sudden, I'm like 242 00:13:58,396 --> 00:14:00,676 Speaker 1: striking guys out on three fastballs in Double A, and 243 00:14:00,716 --> 00:14:03,636 Speaker 1: I'm going like, is it this easy? Like this is 244 00:14:03,676 --> 00:14:06,836 Speaker 1: all it took. By the end of that season, Matt 245 00:14:06,836 --> 00:14:08,876 Speaker 1: Boyd had been called up to the major leagues to 246 00:14:08,956 --> 00:14:12,516 Speaker 1: the Detroit Tigers to be a starting pitcher. Paints in 247 00:14:12,676 --> 00:14:16,876 Speaker 1: and then paints away both fastballs. Want to pitch to 248 00:14:17,076 --> 00:14:23,076 Speaker 1: Bird another strikeout for Matthew boy to start the form. 249 00:14:23,116 --> 00:14:25,236 Speaker 1: They now pay him five point three million dollars a 250 00:14:25,276 --> 00:14:29,076 Speaker 1: year and feel like they're getting a deal. And Kyle Body, well, 251 00:14:29,116 --> 00:14:31,156 Speaker 1: now he has a new job too. I'm the president 252 00:14:31,156 --> 00:14:33,436 Speaker 1: and founder of drive Land Baseball and the director of 253 00:14:33,436 --> 00:14:41,956 Speaker 1: Pitching Initiatives of the Cincinnati Reds. Somebody asks a question, 254 00:14:42,476 --> 00:14:46,276 Speaker 1: how do you throw a ball faster? They gathered data. 255 00:14:46,556 --> 00:14:48,996 Speaker 1: They measure everything that the human body does when it 256 00:14:49,036 --> 00:14:52,756 Speaker 1: throws a ball. They test theories and find answers rooted 257 00:14:52,836 --> 00:14:55,756 Speaker 1: in science. All of a sudden, there's a new way 258 00:14:55,796 --> 00:15:00,036 Speaker 1: to coach, and it's getting adopted in sports. But not 259 00:15:00,196 --> 00:15:03,556 Speaker 1: just in sports, because there are people all over who 260 00:15:03,556 --> 00:15:05,636 Speaker 1: don't know why they're good at something or how to 261 00:15:05,676 --> 00:15:08,876 Speaker 1: get better. About five years ago, I was CEO and 262 00:15:09,116 --> 00:15:13,876 Speaker 1: at a software company and we're growing pretty quickly. This 263 00:15:14,076 --> 00:15:16,956 Speaker 1: is a meet bendoff. Five years ago, he was just 264 00:15:16,996 --> 00:15:20,396 Speaker 1: another Silicon Valley entrepreneur trying to get his product out 265 00:15:20,396 --> 00:15:25,236 Speaker 1: the door by using salespeople. But I was puzzled why 266 00:15:27,036 --> 00:15:30,996 Speaker 1: some of our people were more successful than others. And 267 00:15:31,076 --> 00:15:34,916 Speaker 1: every time we wanted to understand why, we had to 268 00:15:34,956 --> 00:15:40,036 Speaker 1: go and interview people and see what they think. And 269 00:15:40,196 --> 00:15:45,036 Speaker 1: he didn't want just a collection of stories. He wanted data. 270 00:15:45,196 --> 00:15:51,116 Speaker 1: Why did some sales pitches work while others didn't? Think 271 00:15:51,116 --> 00:15:56,436 Speaker 1: about like football right or baseball right? If the coach 272 00:15:56,876 --> 00:15:59,876 Speaker 1: never sees the game and the only thing is to 273 00:16:00,036 --> 00:16:02,076 Speaker 1: understand how to get better is by interviewing some of 274 00:16:02,076 --> 00:16:06,036 Speaker 1: the players what they think has happened. So I wanted 275 00:16:06,076 --> 00:16:08,196 Speaker 1: to have like something like a game tape and game 276 00:16:08,236 --> 00:16:13,676 Speaker 1: stats for sales and AI to understand what really separates 277 00:16:13,676 --> 00:16:17,396 Speaker 1: the top performers from everybody else. You know that message 278 00:16:17,436 --> 00:16:20,716 Speaker 1: that you get on customer service calls, This call may 279 00:16:20,756 --> 00:16:24,876 Speaker 1: be recorded for quality assurance. We appreciate your patience. Calls 280 00:16:25,436 --> 00:16:28,476 Speaker 1: for quality. Companies were recording their sales calls, but no 281 00:16:28,476 --> 00:16:30,916 Speaker 1: one was really listening to them. A met wanted to 282 00:16:30,956 --> 00:16:33,756 Speaker 1: listen to all the game tape and analyze it. And 283 00:16:33,876 --> 00:16:36,676 Speaker 1: I then started asking a bunch of other people, you know, 284 00:16:36,716 --> 00:16:38,716 Speaker 1: if we build a system kind of shine a light 285 00:16:38,796 --> 00:16:41,396 Speaker 1: on conversation and give you insights from that. Would you 286 00:16:41,396 --> 00:16:44,036 Speaker 1: buy how much you're willing to pay a meat? Created 287 00:16:44,076 --> 00:16:47,716 Speaker 1: a new company he called it Gong. He went to 288 00:16:47,756 --> 00:16:51,036 Speaker 1: people and said, hand me all the recordings of all 289 00:16:51,116 --> 00:16:54,356 Speaker 1: your sales calls, plus a list of your salespeople in 290 00:16:54,476 --> 00:16:59,316 Speaker 1: order of how good they are. We'll analyze it. And 291 00:16:59,396 --> 00:17:01,596 Speaker 1: so if you'd gone to just one of those top 292 00:17:01,596 --> 00:17:05,396 Speaker 1: sales people and ask what you're doing, it works. No, no, 293 00:17:05,716 --> 00:17:10,076 Speaker 1: absolutely not, because they don't know. People think it's an art, right. 294 00:17:10,196 --> 00:17:12,756 Speaker 1: It's like they're not aware that's something that they're doing 295 00:17:12,796 --> 00:17:14,356 Speaker 1: because they don't know what the other people are doing, 296 00:17:14,356 --> 00:17:16,836 Speaker 1: so they don't know what the differences are. The Gong 297 00:17:17,036 --> 00:17:20,316 Speaker 1: artificial intelligence, had no theory about what worked and what 298 00:17:20,436 --> 00:17:24,156 Speaker 1: didn't in sales. It just had millions of sales pitches. 299 00:17:24,716 --> 00:17:27,636 Speaker 1: It searched for patterns in both the calls they got 300 00:17:27,676 --> 00:17:32,876 Speaker 1: results and the calls that didn't. The Gong actually learns 301 00:17:32,876 --> 00:17:35,596 Speaker 1: like it looks at the salespeople and says, like, which 302 00:17:35,596 --> 00:17:38,596 Speaker 1: one is closing more deals, and then it starts to 303 00:17:38,596 --> 00:17:41,276 Speaker 1: analyze the difference. So the software learns automatically. You just 304 00:17:41,476 --> 00:17:46,596 Speaker 1: connect to the calls. If there was an art to 305 00:17:46,676 --> 00:17:48,956 Speaker 1: any of this, it was in the questions that Gong 306 00:17:49,116 --> 00:17:51,476 Speaker 1: asked if the phone call data. I mean, the very 307 00:17:51,516 --> 00:17:54,676 Speaker 1: simple one is like percent of talk time? Right, You 308 00:17:54,716 --> 00:17:57,356 Speaker 1: and I are talking right now, and by the end 309 00:17:57,396 --> 00:17:59,356 Speaker 1: of the call, Gong say, well, a meat was talking 310 00:17:59,396 --> 00:18:03,236 Speaker 1: fifty six percent of the time. It turns out there's 311 00:18:03,316 --> 00:18:05,916 Speaker 1: lots of things that separated great sales pitch from a 312 00:18:05,956 --> 00:18:09,716 Speaker 1: bad one. The simple virtues. They're sort of obvious, but 313 00:18:09,796 --> 00:18:13,796 Speaker 1: they could now be quantified. On average, forty six percent 314 00:18:14,396 --> 00:18:21,436 Speaker 1: talk time is ideal. Thirteen questions is optimal? Right? Not less? 315 00:18:21,836 --> 00:18:25,636 Speaker 1: No more so, this alone is food for thought that 316 00:18:25,756 --> 00:18:28,996 Speaker 1: there's a maximum amount of time to be talking. If 317 00:18:29,036 --> 00:18:31,356 Speaker 1: I were to talk only forty six percent of the time, 318 00:18:31,956 --> 00:18:34,076 Speaker 1: would my wife stop telling me that I don't listen? 319 00:18:34,796 --> 00:18:36,916 Speaker 1: What if I counted the number of questions I asked 320 00:18:36,916 --> 00:18:40,956 Speaker 1: in a particular conversation and stopped myself at thirteen. I mean, 321 00:18:40,996 --> 00:18:43,556 Speaker 1: it's not that if you ask fourteen you lose that deal. 322 00:18:43,676 --> 00:18:47,916 Speaker 1: But more than that, people might lose their patience, and 323 00:18:48,796 --> 00:18:53,116 Speaker 1: too less it means you're talking too much. Patience factor, 324 00:18:53,116 --> 00:18:55,276 Speaker 1: that's like a pretty big challenge for a lot of people, 325 00:18:55,396 --> 00:19:00,236 Speaker 1: they're too quick to respond, so it's a good practice 326 00:19:00,276 --> 00:19:03,996 Speaker 1: to pause think in any reply. We all know that 327 00:19:04,716 --> 00:19:07,196 Speaker 1: in theory, at least, Gone could act just like a 328 00:19:07,276 --> 00:19:11,676 Speaker 1: game coach, just last instructions into a person's ear as 329 00:19:11,716 --> 00:19:15,596 Speaker 1: they pitched the product. Shut the hell up, ask another question. 330 00:19:16,036 --> 00:19:19,516 Speaker 1: You're at seventy three percent, and Meek decided it was 331 00:19:19,556 --> 00:19:21,356 Speaker 1: better if he didn't do this. If he offered the 332 00:19:21,396 --> 00:19:25,156 Speaker 1: feedback after the call instead, you know, will alert the 333 00:19:25,276 --> 00:19:32,316 Speaker 1: coaches to hear like three conversations that have room for improvement, 334 00:19:32,476 --> 00:19:35,996 Speaker 1: and then the coach will open them up just like 335 00:19:36,036 --> 00:19:40,196 Speaker 1: a game tape and start breaking it down to start commenting, Oh, 336 00:19:40,516 --> 00:19:43,116 Speaker 1: minute three ZH six, I really love the way you 337 00:19:43,196 --> 00:19:47,356 Speaker 1: phrased that question. Have you considered X man? If I 338 00:19:47,396 --> 00:19:49,676 Speaker 1: were in sales, all of this would drive me batshit, 339 00:19:50,636 --> 00:19:52,876 Speaker 1: which is why I'm not in sales and also probably 340 00:19:52,916 --> 00:19:55,316 Speaker 1: why everyone but my mother is still making fun of 341 00:19:55,356 --> 00:19:59,036 Speaker 1: my podcast ad Reads. But for people who actually make 342 00:19:59,076 --> 00:20:03,396 Speaker 1: their living selling stuff, well, Gong is their new best coach. 343 00:20:03,756 --> 00:20:07,076 Speaker 1: So when I'm engaging with a customer, especially in introductory calls, 344 00:20:07,676 --> 00:20:09,476 Speaker 1: I want them to do as much of the talking 345 00:20:09,556 --> 00:20:12,876 Speaker 1: is possible. That's Megan Dorner, who sells software for a 346 00:20:12,916 --> 00:20:16,196 Speaker 1: Canadian company called AVIC. I have no idea what the 347 00:20:16,196 --> 00:20:18,876 Speaker 1: software it does, but that's not important. The key fact 348 00:20:18,876 --> 00:20:21,156 Speaker 1: about it is that it needs to be sold. The 349 00:20:21,196 --> 00:20:24,516 Speaker 1: more the customers talking, the more that I'm learning, especially 350 00:20:24,516 --> 00:20:27,516 Speaker 1: like I said on that first call. And so the 351 00:20:27,596 --> 00:20:31,516 Speaker 1: longest customer story again is how long they're telling a 352 00:20:32,036 --> 00:20:35,476 Speaker 1: you know, an interactive I guess one long tail. So 353 00:20:35,556 --> 00:20:39,436 Speaker 1: that the customers monologuing, that's good, that's correct, that's correct. 354 00:20:39,516 --> 00:20:43,436 Speaker 1: But if you're monologuing, it's bad. Megan now pays attention 355 00:20:43,436 --> 00:20:49,716 Speaker 1: to who's monologuing, because Gong forces her too. Five minutes 356 00:20:49,756 --> 00:20:52,556 Speaker 1: after every call, she gets a note from Gong filled 357 00:20:52,556 --> 00:20:56,116 Speaker 1: with stats and a color coded report card. One of 358 00:20:56,116 --> 00:20:58,396 Speaker 1: the grades is based on the length of her monologues. 359 00:20:58,676 --> 00:21:02,436 Speaker 1: When the customer monologues, Gong doesn't call it that, They 360 00:21:02,476 --> 00:21:07,396 Speaker 1: call it a story. If you've got a woman of 361 00:21:07,516 --> 00:21:09,196 Speaker 1: few words on the other end of the line and 362 00:21:09,436 --> 00:21:11,236 Speaker 1: there's not much you can do to tease a big 363 00:21:11,236 --> 00:21:13,796 Speaker 1: story out of her, you totally right. And that's when 364 00:21:13,836 --> 00:21:18,316 Speaker 1: I find that my longest monologue, longest customer story, and 365 00:21:18,396 --> 00:21:23,276 Speaker 1: talk ratio could be all in yellow. Yellow is Gong's 366 00:21:23,356 --> 00:21:28,156 Speaker 1: color grade for a C. Green is an A red 367 00:21:28,396 --> 00:21:31,716 Speaker 1: is an F. Megan's never in the red. She's one 368 00:21:31,716 --> 00:21:35,476 Speaker 1: of Avic's top salespeople and also almost by definition, a 369 00:21:35,596 --> 00:21:39,596 Speaker 1: salesperson that Gong thinks is great. The only yellow card 370 00:21:39,636 --> 00:21:41,956 Speaker 1: she's ever likely to get from Gong is for interrupting. 371 00:21:42,516 --> 00:21:45,636 Speaker 1: I'm somebody who likes other people. I like talking, I 372 00:21:45,796 --> 00:21:49,356 Speaker 1: like engaging. I also like to be right. I like 373 00:21:49,556 --> 00:21:54,316 Speaker 1: to help. I like to give the information I think 374 00:21:54,396 --> 00:21:56,316 Speaker 1: is going to be helpful, and so I'm eager to 375 00:21:56,356 --> 00:21:58,476 Speaker 1: do that. So, in a funny way, you had to 376 00:21:58,476 --> 00:22:00,836 Speaker 1: be a different kind of person when you were a 377 00:22:00,836 --> 00:22:04,876 Speaker 1: salesperson than you were just out in the world. That's correct. 378 00:22:08,476 --> 00:22:11,716 Speaker 1: All this lead It's to an obvious thought. If Gone's 379 00:22:11,756 --> 00:22:19,756 Speaker 1: report cards work for sales, why stop there? So after 380 00:22:20,556 --> 00:22:23,276 Speaker 1: I'll be with my friends or with family, I'll think 381 00:22:23,316 --> 00:22:25,836 Speaker 1: to myself, after I leave, God, I interrupted a lot, 382 00:22:26,436 --> 00:22:28,716 Speaker 1: and it's just it's always in the back of my 383 00:22:28,756 --> 00:22:30,796 Speaker 1: mind now because I see it so often that my 384 00:22:30,836 --> 00:22:33,916 Speaker 1: patience is low. Is it Gone that made you aware 385 00:22:33,956 --> 00:22:38,716 Speaker 1: of it more so than before? Absolutely? Does anybody around 386 00:22:38,756 --> 00:22:41,876 Speaker 1: you sense that your approach is changing at all? Is 387 00:22:41,916 --> 00:22:44,836 Speaker 1: anybody noted to you that you know, you're all of 388 00:22:44,876 --> 00:22:49,036 Speaker 1: a sudden listening better to me, my spousehays does that count? 389 00:22:49,596 --> 00:22:53,116 Speaker 1: Oh my God, tell me about that. Yes. So I 390 00:22:53,996 --> 00:22:56,356 Speaker 1: We've had the conversation after we've left a social setting 391 00:22:56,716 --> 00:23:00,476 Speaker 1: where I've said, god, I did I was interrupting a 392 00:23:00,476 --> 00:23:04,316 Speaker 1: lot at dinner or whatever, and she has said to me, 393 00:23:04,676 --> 00:23:08,996 Speaker 1: you know you've been actually better than you have been previous. 394 00:23:09,276 --> 00:23:11,196 Speaker 1: You know you're getting better at listening, You're getting better 395 00:23:11,236 --> 00:23:15,756 Speaker 1: at not interrupting. So she has noticed that, Yeah, I don't, 396 00:23:15,876 --> 00:23:18,636 Speaker 1: I don't do it quite as often. Did you then 397 00:23:18,716 --> 00:23:21,996 Speaker 1: explain to her why you weren't doing it as often? No, 398 00:23:22,156 --> 00:23:26,956 Speaker 1: absolutely not. Life is a market. Everyone's always looking for 399 00:23:27,076 --> 00:23:29,996 Speaker 1: edges that other people don't have. Even a nice Canadian 400 00:23:30,036 --> 00:23:33,396 Speaker 1: woman knows better than to give her secrets away. Do 401 00:23:33,436 --> 00:23:35,276 Speaker 1: you think the world would be a better place if 402 00:23:35,316 --> 00:23:40,236 Speaker 1: we were all gone in all of our conversations. I 403 00:23:40,276 --> 00:23:43,276 Speaker 1: think it depends on the level of you know, give 404 00:23:43,276 --> 00:23:45,676 Speaker 1: a shit that the person has to want to do 405 00:23:45,796 --> 00:23:48,836 Speaker 1: better or be better, right. But I think a lot 406 00:23:48,876 --> 00:23:51,036 Speaker 1: of people who spend a lot of time just talking 407 00:23:51,036 --> 00:23:55,276 Speaker 1: about themselves are unaware how much time they're spending talking 408 00:23:55,316 --> 00:23:59,516 Speaker 1: about themselves. That's very true. Yeah, there's no question that 409 00:23:59,596 --> 00:24:02,596 Speaker 1: the future of this is gone trying to figure out 410 00:24:03,756 --> 00:24:08,156 Speaker 1: and reduced to little color coded bars. What it is 411 00:24:08,196 --> 00:24:12,636 Speaker 1: that causes a uson to like another person? Oh yeah, 412 00:24:12,676 --> 00:24:16,476 Speaker 1: which is scary when you think about it, but incredibly useful. 413 00:24:17,556 --> 00:24:21,076 Speaker 1: Gong's now being used by fifty thousand salespeople at almost 414 00:24:21,076 --> 00:24:24,756 Speaker 1: a thousand different companies. We grab Megan sort of at random, 415 00:24:25,036 --> 00:24:27,556 Speaker 1: and she's a sample size of one. But it's not 416 00:24:27,596 --> 00:24:30,476 Speaker 1: hard to see how this new coaching tool might shape behavior. 417 00:24:30,996 --> 00:24:35,116 Speaker 1: It creates new stats, Gong stats. The stats capture something 418 00:24:35,196 --> 00:24:37,436 Speaker 1: true about the performance, the same way that I don't 419 00:24:37,436 --> 00:24:40,516 Speaker 1: know on base percentage captures something true about a baseball 420 00:24:40,556 --> 00:24:43,716 Speaker 1: hitter's performance. The management pays more to the people with 421 00:24:43,756 --> 00:24:47,356 Speaker 1: good stats, and so soon everyone's just adjusting their game. 422 00:24:48,196 --> 00:24:51,836 Speaker 1: Baseball hitters are learning plate discipline, and salespeople are learning patients. 423 00:24:52,276 --> 00:24:55,516 Speaker 1: Most salespeople would say like Gong is like change my 424 00:24:55,676 --> 00:24:58,756 Speaker 1: game forever. This is a meet bendoff. Again, they use 425 00:24:58,836 --> 00:25:02,356 Speaker 1: those were like a game changer. It is possible if 426 00:25:03,436 --> 00:25:06,956 Speaker 1: if your product isn't good or a competitive in the market, 427 00:25:07,156 --> 00:25:09,956 Speaker 1: it doesn't matter if he has like thirds question, right. 428 00:25:09,956 --> 00:25:12,036 Speaker 1: I mean, it's like, so there's only so much that 429 00:25:12,076 --> 00:25:14,396 Speaker 1: you can do. I mean there's like, uh, this is 430 00:25:14,436 --> 00:25:17,756 Speaker 1: not some kind of magic, it's just like facts. But 431 00:25:17,836 --> 00:25:19,956 Speaker 1: there's something else that happens with data. In the right 432 00:25:19,996 --> 00:25:23,396 Speaker 1: analysis of it, it causes all kinds of folk beliefs 433 00:25:23,436 --> 00:25:26,196 Speaker 1: to just disappear. Let me give you an anecdote. So, 434 00:25:26,236 --> 00:25:30,036 Speaker 1: a lot of the managers and the coaches are often 435 00:25:30,076 --> 00:25:33,316 Speaker 1: obsessed with the filler words. When we introduce the product, Oh, 436 00:25:33,316 --> 00:25:35,396 Speaker 1: can you track like fuller words, you know, like oombs 437 00:25:35,436 --> 00:25:39,396 Speaker 1: and arms and like and so and like. It drives 438 00:25:39,396 --> 00:25:43,396 Speaker 1: them nuts. But you know what, we ran the research 439 00:25:44,276 --> 00:25:48,556 Speaker 1: and turns out there's zero correlation between like usage of 440 00:25:48,996 --> 00:25:54,556 Speaker 1: filler words and success. Huh, So, so that parameter doesn't 441 00:25:54,556 --> 00:25:57,476 Speaker 1: really matter. The managers were thinking that the more filler 442 00:25:57,556 --> 00:26:00,716 Speaker 1: words the worse. Yes, right, it is annoying to hear 443 00:26:00,756 --> 00:26:03,396 Speaker 1: them when you listen to recording and there's and people 444 00:26:03,476 --> 00:26:07,836 Speaker 1: say like a hundred times it drives you nuts, right right. 445 00:26:07,996 --> 00:26:11,156 Speaker 1: The fact is it doesn't matter. The fact is you 446 00:26:11,276 --> 00:26:14,316 Speaker 1: never know any of this without all these facts. Gong 447 00:26:14,476 --> 00:26:19,036 Speaker 1: is screwing up everybody's assumptions. Like the recent one was 448 00:26:19,156 --> 00:26:24,276 Speaker 1: using swear words on calls, does it help, does it 449 00:26:24,316 --> 00:26:27,316 Speaker 1: get in a way or does it even matter? And 450 00:26:27,316 --> 00:26:31,876 Speaker 1: what's the answer. Um, So it depends what the correlation, 451 00:26:32,116 --> 00:26:35,036 Speaker 1: what the research shows. I said, it depends on what 452 00:26:35,076 --> 00:26:39,156 Speaker 1: you're selling, Like if you're selling bibles, it probably doesn't help. Yeah, yeah, 453 00:26:39,196 --> 00:26:42,956 Speaker 1: we didn't have any Bible sellers over there in the stats. 454 00:26:42,996 --> 00:26:46,036 Speaker 1: But it's run over like, you know, fifty thousand people, 455 00:26:46,116 --> 00:26:51,036 Speaker 1: so it's it's a very large number. But the corollary 456 00:26:51,396 --> 00:26:55,036 Speaker 1: is that it depends who starts first. If the buyer 457 00:26:55,396 --> 00:26:59,676 Speaker 1: starts with curse words and you match that that that's 458 00:26:59,676 --> 00:27:03,716 Speaker 1: actually works better. But if you start it it doesn't help. 459 00:27:04,316 --> 00:27:06,556 Speaker 1: You might fucking believe that, or you might fucking not. 460 00:27:07,036 --> 00:27:09,316 Speaker 1: The point is that it doesn't matter what you believe. 461 00:27:10,076 --> 00:27:13,356 Speaker 1: The data generates the knowledge, and the knowledge allows you 462 00:27:13,396 --> 00:27:15,676 Speaker 1: to coach people how to do better at the most 463 00:27:15,676 --> 00:27:23,556 Speaker 1: basic human activity talking. Can you imagine a future like 464 00:27:23,996 --> 00:27:28,356 Speaker 1: that distant future where human interactions will have changed pretty 465 00:27:28,396 --> 00:27:31,556 Speaker 1: meaningfully because we'll all have been coached up in how 466 00:27:31,596 --> 00:27:36,836 Speaker 1: to have conversations. Absolutely, the device that we're using today, 467 00:27:37,116 --> 00:27:41,236 Speaker 1: like namely like speech or the English language hasn't changed 468 00:27:41,476 --> 00:27:45,316 Speaker 1: much for thousands of years, where the amount of information 469 00:27:45,356 --> 00:27:47,796 Speaker 1: that we need to exchange today and the amount of 470 00:27:47,876 --> 00:27:51,676 Speaker 1: noise and clutter and the environment is so different that 471 00:27:51,876 --> 00:27:57,316 Speaker 1: it could definitely use an upgrade. An upgrade. Well, one 472 00:27:57,356 --> 00:27:59,996 Speaker 1: thing hasn't changed. If they're better ways to manipulate people, 473 00:28:00,516 --> 00:28:02,916 Speaker 1: salespeople will find them first and the rest of us 474 00:28:02,916 --> 00:28:06,836 Speaker 1: will just follow. We focus on things that are teachable, 475 00:28:07,076 --> 00:28:09,476 Speaker 1: are coachable, all right, we can't teach anybody how to 476 00:28:09,556 --> 00:28:11,596 Speaker 1: be funny. Let me stop, Let me stop, let me 477 00:28:11,596 --> 00:28:14,956 Speaker 1: stop you for a second. I'm sorry, I know I interrupted, 478 00:28:15,236 --> 00:28:18,076 Speaker 1: but this is important. You don't think you could teach 479 00:28:18,076 --> 00:28:20,796 Speaker 1: people to be funny. Well, we haven't been able to 480 00:28:20,836 --> 00:28:24,716 Speaker 1: crack that code yet. Maybe a meat hasn't cracked that code, 481 00:28:25,156 --> 00:28:36,396 Speaker 1: But that doesn't mean that it can't be cracked. Well, 482 00:28:36,396 --> 00:28:39,396 Speaker 1: you can measure precise things contained in the conversation, the 483 00:28:39,436 --> 00:28:42,636 Speaker 1: words that people are saying to each other, and how 484 00:28:42,636 --> 00:28:46,876 Speaker 1: it influences outcomes. Allison wood Brooks is an associate professor 485 00:28:46,876 --> 00:28:49,316 Speaker 1: at the Harvard Business School. Like, we're on a date, 486 00:28:49,356 --> 00:28:50,916 Speaker 1: do you want to go out with me again? We're 487 00:28:50,916 --> 00:28:53,116 Speaker 1: on a sales call? Did it convert to a sale? 488 00:28:53,796 --> 00:28:56,236 Speaker 1: All kinds of things that you can connect the content 489 00:28:56,276 --> 00:28:59,076 Speaker 1: of the conversation with things that really matter, with outcomes 490 00:28:59,076 --> 00:29:02,716 Speaker 1: that really matter. Professor Brooks takes all these conversations and 491 00:29:02,836 --> 00:29:06,396 Speaker 1: analyzes them. She's using the same new machine learning that 492 00:29:06,476 --> 00:29:09,756 Speaker 1: Gong uses, but she's looking for different things. Give me 493 00:29:09,796 --> 00:29:12,996 Speaker 1: an example, one example of something that you can you 494 00:29:13,116 --> 00:29:16,196 Speaker 1: learn from this technology, Like something someone says that leads 495 00:29:16,236 --> 00:29:18,356 Speaker 1: the other person to want to go on a date. 496 00:29:18,836 --> 00:29:22,316 Speaker 1: Let's give you that exact example. We have a data 497 00:29:22,316 --> 00:29:25,436 Speaker 1: set of people doing speed dating. Each person went on 498 00:29:25,476 --> 00:29:28,636 Speaker 1: like twenty dates, okay, quick four to five minute speed 499 00:29:28,716 --> 00:29:32,316 Speaker 1: dates in round robin fashion. At the end of each date, 500 00:29:32,356 --> 00:29:33,796 Speaker 1: you say are you willing to go out with that 501 00:29:33,876 --> 00:29:37,956 Speaker 1: person again? And they record and transcribe all the interactions. 502 00:29:38,236 --> 00:29:40,556 Speaker 1: So now we see exactly what people are saying on 503 00:29:40,556 --> 00:29:42,316 Speaker 1: their dates, and we can measure what are the things 504 00:29:42,356 --> 00:29:44,516 Speaker 1: that people are doing and saying that make them more 505 00:29:44,556 --> 00:29:47,956 Speaker 1: likely to be more datable in the future. And one 506 00:29:47,956 --> 00:29:50,796 Speaker 1: thing that really matters is question asking. So asking more 507 00:29:50,876 --> 00:29:54,156 Speaker 1: questions for both men and women on these heterosexual dates 508 00:29:54,236 --> 00:29:58,996 Speaker 1: leads to better dating outcomes. Really, especially follow up questions. 509 00:29:59,356 --> 00:30:01,116 Speaker 1: How do you know that the people who are asking 510 00:30:01,116 --> 00:30:03,276 Speaker 1: the questions aren't just naturally more attracted to the people 511 00:30:03,276 --> 00:30:05,956 Speaker 1: who they're asking the questions of. Because so two things. 512 00:30:06,036 --> 00:30:09,956 Speaker 1: One we can control for other aspects of attractiveness observational 513 00:30:10,036 --> 00:30:13,116 Speaker 1: data too. We then come back to Harvard and run 514 00:30:13,156 --> 00:30:16,636 Speaker 1: experiments where we tell people ask a lot of questions 515 00:30:16,636 --> 00:30:18,916 Speaker 1: in one condition or another condition where we say we 516 00:30:18,916 --> 00:30:22,036 Speaker 1: don't tell them anything. In the condition where they're asking 517 00:30:22,036 --> 00:30:24,916 Speaker 1: a lot of questions, they also are more attractive and likable. 518 00:30:25,636 --> 00:30:28,556 Speaker 1: This new line of research has uncovered the various ways 519 00:30:28,596 --> 00:30:32,956 Speaker 1: that people gain power and authority in conversation. This professor 520 00:30:32,996 --> 00:30:35,396 Speaker 1: can prove that they work, and it's just made her 521 00:30:35,476 --> 00:30:39,116 Speaker 1: want to ask even more questions. Even if you know 522 00:30:40,156 --> 00:30:45,836 Speaker 1: conceptually what a charismatic, smooth, productive conversationalist looks like. Is 523 00:30:45,876 --> 00:30:48,756 Speaker 1: there any way that you can train people to actually 524 00:30:49,036 --> 00:30:52,676 Speaker 1: get better at executing it? Brooks has all this data, 525 00:30:52,796 --> 00:30:55,796 Speaker 1: some of it from Gong, but also from doctor's appointments 526 00:30:55,796 --> 00:30:58,996 Speaker 1: and work meetings and parole interviews and speed dating sessions 527 00:30:58,996 --> 00:31:01,596 Speaker 1: and on and on and on. She uses the data 528 00:31:01,676 --> 00:31:05,556 Speaker 1: to test theories about conversations, and one of those theories 529 00:31:05,676 --> 00:31:09,236 Speaker 1: is about humor. There's this great work that sends of 530 00:31:09,356 --> 00:31:15,396 Speaker 1: play right is like the key to psychological safety and 531 00:31:15,956 --> 00:31:19,516 Speaker 1: thriving and creativity. It's the only way that you can 532 00:31:19,516 --> 00:31:21,636 Speaker 1: really be creative in the presence of others is if 533 00:31:21,636 --> 00:31:24,556 Speaker 1: you feel safe to say something stupid and silly. That 534 00:31:24,636 --> 00:31:26,676 Speaker 1: all sounds sensible to me, But do you have any 535 00:31:27,756 --> 00:31:30,876 Speaker 1: data to back up absolutely? What's the data? Let me 536 00:31:30,876 --> 00:31:33,996 Speaker 1: tell you about one paper. It's a bidirectional finding, meaning 537 00:31:34,316 --> 00:31:38,596 Speaker 1: people who have high status tend to be more free 538 00:31:38,636 --> 00:31:41,396 Speaker 1: and use humor more freely. But the more interesting direction 539 00:31:41,556 --> 00:31:44,956 Speaker 1: is if you are of low status, if you can 540 00:31:45,076 --> 00:31:49,116 Speaker 1: land a joke, people perceive you as higher status. A 541 00:31:49,156 --> 00:31:51,916 Speaker 1: lot of polite laughter happens. So but if other people 542 00:31:51,956 --> 00:31:55,676 Speaker 1: think what you said is actually funny, appropriate for the circumstances, 543 00:31:55,796 --> 00:31:58,956 Speaker 1: and at least one person laughs at it, your status 544 00:31:58,996 --> 00:32:05,036 Speaker 1: takes this huge jump. Totally true, right, So yes, it 545 00:32:05,116 --> 00:32:08,316 Speaker 1: pays to be funny. Funny gets you status, and status 546 00:32:08,316 --> 00:32:13,036 Speaker 1: get you money. But the mystery remains, can funny be coached? 547 00:32:14,156 --> 00:32:16,636 Speaker 1: There's a part of me that wonders if there's a 548 00:32:16,756 --> 00:32:20,876 Speaker 1: version of what's going on in baseball coaching that doesn't 549 00:32:20,916 --> 00:32:23,516 Speaker 1: apply to what you're doing. And what's going on baseball 550 00:32:23,556 --> 00:32:27,716 Speaker 1: coaching is the technology has generated all this data about 551 00:32:27,716 --> 00:32:32,556 Speaker 1: how a pitcher's body moves, and they're able to identify 552 00:32:33,156 --> 00:32:36,796 Speaker 1: people who have an aptitude for doing things that is 553 00:32:36,876 --> 00:32:40,556 Speaker 1: untapped because they have whatever. The fundamental attribute is arm 554 00:32:40,596 --> 00:32:44,956 Speaker 1: speed that you can't teach, but translating the arm speed 555 00:32:44,996 --> 00:32:47,196 Speaker 1: into a speed on a fastball is a different thing. 556 00:32:47,556 --> 00:32:50,276 Speaker 1: And I wonder if there's an equivalent in conversation where 557 00:32:50,956 --> 00:32:56,676 Speaker 1: you could identify the core traits that lead to conversational 558 00:32:56,876 --> 00:33:00,596 Speaker 1: excellence and you can figure out who's sort of maxing 559 00:33:00,596 --> 00:33:06,916 Speaker 1: out and who's not and why totally. And not only that, 560 00:33:07,636 --> 00:33:10,716 Speaker 1: Allison's taken everything she's learned about human speech, humor, and 561 00:33:10,756 --> 00:33:13,116 Speaker 1: the rest and built it into a new course at 562 00:33:13,116 --> 00:33:16,676 Speaker 1: the Harvard Business School. It's called how to Talk Gooder 563 00:33:17,156 --> 00:33:20,356 Speaker 1: in Business and in Life. It's trying to turn conversation 564 00:33:20,556 --> 00:33:23,436 Speaker 1: into a science. It's like having the right arm for 565 00:33:23,476 --> 00:33:26,316 Speaker 1: the fastball and not doing it quite right. You've got 566 00:33:26,396 --> 00:33:28,556 Speaker 1: the brain space and the ability to do it, you 567 00:33:28,596 --> 00:33:31,716 Speaker 1: just didn't think to do it. That's the dream scenario, 568 00:33:31,756 --> 00:33:35,196 Speaker 1: and that's really the hope with these HBS students, right 569 00:33:35,236 --> 00:33:39,476 Speaker 1: These are super smart people who maybe just haven't heard 570 00:33:39,516 --> 00:33:43,076 Speaker 1: the right strategies. Hundreds of Harvard students tried to get 571 00:33:43,116 --> 00:33:47,236 Speaker 1: into Alison's new class. She accepted seventy seventy of the 572 00:33:47,276 --> 00:33:51,236 Speaker 1: world's most ambitious people, hoping that the science of conversation 573 00:33:51,676 --> 00:33:55,316 Speaker 1: will offer them yet another edge in life. Many people 574 00:33:55,356 --> 00:33:58,516 Speaker 1: would find it odd to approach a conversation seeking to 575 00:33:58,596 --> 00:34:03,196 Speaker 1: maximize the profits of it. Exactly, They'll love it. They 576 00:34:03,196 --> 00:34:06,156 Speaker 1: want to perform optimally in every way in their lives, 577 00:34:06,436 --> 00:34:08,796 Speaker 1: and this is the moment to moment way that you 578 00:34:08,836 --> 00:34:13,196 Speaker 1: would achieve that moment to moment, and in each moment, 579 00:34:13,236 --> 00:34:15,876 Speaker 1: there's now data which can be used by a coach. 580 00:34:16,316 --> 00:34:19,676 Speaker 1: I'll be checking in with Alison's students next week and 581 00:34:19,796 --> 00:34:22,996 Speaker 1: asking a new question. When you start to get an 582 00:34:23,076 --> 00:34:27,436 Speaker 1: edge like this, a data edge, does the coaching start 583 00:34:27,476 --> 00:34:31,916 Speaker 1: to shade into something else, something that's maybe against the rules. 584 00:34:38,396 --> 00:34:41,556 Speaker 1: I'm Michael Lewis, Thanks for listening to Against the Rules. 585 00:34:42,516 --> 00:34:45,076 Speaker 1: Against the Rules is brought to you by Pushkin Industries. 586 00:34:45,276 --> 00:34:49,756 Speaker 1: The show's produced by Audrey Dilling and Catherine girodo Or 587 00:34:49,876 --> 00:34:53,996 Speaker 1: Girardo or girodot Or Girardo I've Never Gotten It right, 588 00:34:54,276 --> 00:34:58,636 Speaker 1: with research assistance from Lydia Genecott and Zooe Wynn Our 589 00:34:58,796 --> 00:35:03,036 Speaker 1: editor is the magnificent Julia Barton, who finds my every mistake. 590 00:35:03,396 --> 00:35:06,676 Speaker 1: Mia Loebell is our executive producer, and she disapproves of 591 00:35:06,716 --> 00:35:10,596 Speaker 1: me half the time. Our theme was composed by Nick Brittell, 592 00:35:10,716 --> 00:35:14,116 Speaker 1: who is really slumming it working here, with additional scoring 593 00:35:14,196 --> 00:35:18,596 Speaker 1: by Stellwagon Symphonette. We got fact checked by Beth Johnson, 594 00:35:18,636 --> 00:35:22,036 Speaker 1: which was totally unnecessary because our facts are always right. 595 00:35:22,356 --> 00:35:25,236 Speaker 1: And our show was recorded by tofur Ruth and Trey 596 00:35:25,316 --> 00:35:29,476 Speaker 1: Schiltz in spite of enduring the coronavirus in the studio, 597 00:35:29,836 --> 00:35:32,836 Speaker 1: which is the Northgate Studios in Berkeley, as always thanks 598 00:35:32,876 --> 00:35:35,996 Speaker 1: to Pushkin's founders Jacob Weisberg, who I think of as 599 00:35:36,076 --> 00:35:38,756 Speaker 1: my brain, and Malcolm Gladwell, who I think of as 600 00:35:38,796 --> 00:35:51,956 Speaker 1: my goal. Do you sense that the challenge with men 601 00:35:52,076 --> 00:35:54,556 Speaker 1: is different from the challenge with women. We do have 602 00:35:54,676 --> 00:36:00,636 Speaker 1: a good amount of research on gender in conversation. For example, 603 00:36:00,836 --> 00:36:03,756 Speaker 1: we know Matthias Melt the University of Arizona has this 604 00:36:03,876 --> 00:36:07,116 Speaker 1: great paper showing that men and women are equally talkative 605 00:36:07,436 --> 00:36:10,156 Speaker 1: and even though women have the stereo type of being chattier, 606 00:36:10,516 --> 00:36:12,596 Speaker 1: what he finds is that men and women both speak 607 00:36:12,636 --> 00:36:16,276 Speaker 1: about sixteen thousand words per day, so that's about similar. 608 00:36:16,316 --> 00:36:18,396 Speaker 1: But what we do know from other researches that men 609 00:36:18,396 --> 00:36:22,436 Speaker 1: and women speak at different times, especially in the workplace. 610 00:36:22,556 --> 00:36:24,996 Speaker 1: You know, I find I speak when a woman needs 611 00:36:24,996 --> 00:36:30,116 Speaker 1: something explained to her, exactly fifteen and a half thousand 612 00:36:30,156 --> 00:36:32,996 Speaker 1: and of my words every day just to do that, 613 00:36:33,316 --> 00:36:37,036 Speaker 1: and for some reason, for some reason, they don't appreciate it. 614 00:36:38,636 --> 00:36:41,316 Speaker 1: This just is Could I learn how to get them 615 00:36:41,316 --> 00:36:45,476 Speaker 1: to appreciate it by coming to your class? No, No, 616 00:36:46,916 --> 00:36:50,236 Speaker 1: what kind of class is this? Oh my god, it's 617 00:36:50,316 --> 00:36:52,916 Speaker 1: too funny.