1 00:00:02,480 --> 00:00:12,360 Speaker 1: This is your host, JJ zach j J Zacherson. What's 2 00:00:12,440 --> 00:00:15,040 Speaker 1: up everyone. It's JJ Zachareson, the editor in chief over 3 00:00:15,080 --> 00:00:18,000 Speaker 1: at number fire dot com and this is episode twenty 4 00:00:18,079 --> 00:00:20,200 Speaker 1: nine of the Late Round podcast, one of the many 5 00:00:20,200 --> 00:00:22,600 Speaker 1: shows that are part of the number fire podcast network. 6 00:00:23,040 --> 00:00:26,320 Speaker 1: Thanks for tuning in. I've got this little dog named Henry. 7 00:00:26,480 --> 00:00:29,280 Speaker 1: He's a fifteen pound multi Shitzuo mix and he's dope 8 00:00:29,320 --> 00:00:32,400 Speaker 1: af He just turns seven, and he has more personality 9 00:00:32,440 --> 00:00:35,040 Speaker 1: than a lot of humans do, especially humans who analyze 10 00:00:35,080 --> 00:00:37,560 Speaker 1: fantasy football for a living. Henry's on a special diet 11 00:00:37,560 --> 00:00:39,920 Speaker 1: because the dude is allergic to literally everything he eats 12 00:00:39,960 --> 00:00:42,800 Speaker 1: this weird mix of duck and potatoes. Now, literally for 13 00:00:42,920 --> 00:00:46,120 Speaker 1: every meal it's duck and potatoes, though I guess for 14 00:00:46,159 --> 00:00:48,640 Speaker 1: a dog it could be a lot worse. Anyway, my 15 00:00:48,680 --> 00:00:50,920 Speaker 1: wife and I get his food delivered to our house 16 00:00:51,000 --> 00:00:52,560 Speaker 1: so that we don't have to go to the vet 17 00:00:52,560 --> 00:00:54,960 Speaker 1: and pick it up every time he runs out, and 18 00:00:55,000 --> 00:00:57,280 Speaker 1: it's on a set schedule because it's pretty easy to 19 00:00:57,320 --> 00:01:00,440 Speaker 1: predict when he'll run out of food, and when he'll 20 00:01:00,480 --> 00:01:02,720 Speaker 1: run out of food even before he runs out of food. 21 00:01:03,080 --> 00:01:06,360 Speaker 1: Is a huge advantage for budgeting purposes and honestly for 22 00:01:06,440 --> 00:01:09,360 Speaker 1: storage purposes because he gets a lot of food when 23 00:01:09,360 --> 00:01:11,760 Speaker 1: it's delivered. There's a lot of duck and potatoes that 24 00:01:11,800 --> 00:01:14,640 Speaker 1: come to our front door when it gets delivered. Despite 25 00:01:14,640 --> 00:01:17,560 Speaker 1: being a small dog, Henry can crush some cans of 26 00:01:17,640 --> 00:01:20,360 Speaker 1: duck and potato. This is a prime example of why 27 00:01:20,400 --> 00:01:24,280 Speaker 1: and how predictability can be super useful and helpful and 28 00:01:24,360 --> 00:01:27,440 Speaker 1: fantasy football. And this is just a personal subjective opinion. 29 00:01:27,840 --> 00:01:31,120 Speaker 1: I think the notion of predictability is really overlooked. I'm 30 00:01:31,120 --> 00:01:33,600 Speaker 1: not talking about I think this guy will finish as 31 00:01:33,640 --> 00:01:35,360 Speaker 1: a wide receiver two at the end of the season. 32 00:01:35,800 --> 00:01:38,320 Speaker 1: I mean predictability in terms of how we're making start 33 00:01:38,360 --> 00:01:42,319 Speaker 1: sit decisions each week. Every day during fantasy football season, 34 00:01:42,640 --> 00:01:46,280 Speaker 1: my Twitter mentions blow up with people asking start sick questions. 35 00:01:46,400 --> 00:01:48,760 Speaker 1: Should I start Frank Gore or Rob Kelly? They ask? 36 00:01:49,120 --> 00:01:51,840 Speaker 1: But the question is never does Frank Gore have more 37 00:01:51,880 --> 00:01:55,880 Speaker 1: predictable output than Rob Kelly does? Or which positions in 38 00:01:55,920 --> 00:01:59,720 Speaker 1: fantasy football are most predictable. I've alluded to this concept 39 00:01:59,800 --> 00:02:02,440 Speaker 1: men many times on previous podcasts, but I figured I'd 40 00:02:02,480 --> 00:02:04,760 Speaker 1: do a show on it because, like I said, I 41 00:02:04,760 --> 00:02:08,000 Speaker 1: think It's an overlooked idea within the community, especially when 42 00:02:08,040 --> 00:02:11,320 Speaker 1: you consider how focused people are on season long projections 43 00:02:11,560 --> 00:02:15,919 Speaker 1: and not how those points are actually accumulated. So predictability, 44 00:02:15,960 --> 00:02:18,359 Speaker 1: as I like to call it, is simply how foreseeable 45 00:02:18,480 --> 00:02:21,240 Speaker 1: a certain performance is. Think of it this way. You're 46 00:02:21,280 --> 00:02:23,440 Speaker 1: setting up a lineup each week, and within your lineup, 47 00:02:23,440 --> 00:02:26,080 Speaker 1: you're asking yourself should I play this guy or this guy. 48 00:02:26,520 --> 00:02:29,320 Speaker 1: It's a lot easier to make those decisions when you're 49 00:02:29,360 --> 00:02:32,919 Speaker 1: confident about how a player is going to perform. Predictability 50 00:02:32,960 --> 00:02:37,440 Speaker 1: isn't just about individual players either, It's about positions. For instance, 51 00:02:37,760 --> 00:02:40,560 Speaker 1: of all four of the big positions in fantasy football, 52 00:02:40,639 --> 00:02:44,960 Speaker 1: so quarterbacks, tight ends, running backs, wide receivers, quarterback is 53 00:02:45,000 --> 00:02:49,200 Speaker 1: by far the most predictable. It makes sense intuitively, they're 54 00:02:49,280 --> 00:02:52,080 Speaker 1: dropping back to pass thirty plus times per game, and 55 00:02:52,160 --> 00:02:54,919 Speaker 1: as a result, you have a larger sample on them 56 00:02:54,960 --> 00:02:58,920 Speaker 1: heading into a particular contest. But moreover, fantasy scoring just 57 00:02:58,960 --> 00:03:01,760 Speaker 1: sort of makes them more didictable because quarterbacks are gaining 58 00:03:01,800 --> 00:03:04,280 Speaker 1: some sort of points in fantasy for almost any action 59 00:03:04,400 --> 00:03:06,400 Speaker 1: they take on the field, and since they make so 60 00:03:06,480 --> 00:03:10,400 Speaker 1: many actions in a game, their floor is fundamentally higher 61 00:03:10,440 --> 00:03:13,919 Speaker 1: than other positions in fantasy. In other words, a quarterback's 62 00:03:14,000 --> 00:03:18,040 Speaker 1: variance from a weekly fantasy output standpoint may be less 63 00:03:18,160 --> 00:03:21,120 Speaker 1: than what we see from other positions. I actually mentioned 64 00:03:21,120 --> 00:03:23,720 Speaker 1: this on last week's Mailbag episode, but there's a statistic 65 00:03:23,760 --> 00:03:26,480 Speaker 1: called coefficient of variation that I've used since I started 66 00:03:26,480 --> 00:03:30,040 Speaker 1: doing fantasy football writing that helps us gauge how predictable 67 00:03:30,400 --> 00:03:32,600 Speaker 1: a set of data is. It tells us how much 68 00:03:32,600 --> 00:03:35,480 Speaker 1: the set of data varies, or the standard deviation divided 69 00:03:35,520 --> 00:03:38,520 Speaker 1: by the set of datas mean or average. The reason 70 00:03:38,520 --> 00:03:40,680 Speaker 1: you want to look at the average and the denominator 71 00:03:40,680 --> 00:03:43,800 Speaker 1: of the equation is because standard deviation alone does not 72 00:03:44,000 --> 00:03:47,600 Speaker 1: tell us enough. For instance, in twenty sixteen, Drew Brees's 73 00:03:47,640 --> 00:03:50,040 Speaker 1: week to week output had a standard deviation of ten 74 00:03:50,040 --> 00:03:53,240 Speaker 1: point seven to one points in standard scoring leagues. Meanwhile, 75 00:03:53,360 --> 00:03:56,480 Speaker 1: Lamar Miller's standard deviation in twenty fourteen was just six 76 00:03:56,520 --> 00:03:59,680 Speaker 1: point five to eight PPR points. So for Drew Brees, 77 00:03:59,760 --> 00:04:02,920 Speaker 1: his Sander deviation was ten point seven one points, Lamar 78 00:04:02,960 --> 00:04:06,040 Speaker 1: Miller's was six point five eight points. This doesn't mean 79 00:04:06,080 --> 00:04:09,320 Speaker 1: that Lamar Miller was a more consistent fantasy asset in 80 00:04:09,360 --> 00:04:13,320 Speaker 1: twenty fourteen than Drew Brees was last year. Since quarterbacks 81 00:04:13,320 --> 00:04:16,720 Speaker 1: score the most points in fantasy football, their overall range 82 00:04:16,720 --> 00:04:19,320 Speaker 1: of scoring is larger. When a set of data is 83 00:04:19,360 --> 00:04:22,840 Speaker 1: more spread out, you end up getting higher standard deviations. 84 00:04:23,040 --> 00:04:26,159 Speaker 1: And that's why the coefficient of variation exists. It takes 85 00:04:26,160 --> 00:04:28,800 Speaker 1: standard deviation and divides it by the data sets mean 86 00:04:29,200 --> 00:04:31,200 Speaker 1: in order to get more of an apples to Apple's 87 00:04:31,240 --> 00:04:34,160 Speaker 1: comparison between different data sets. While Drew Brees in twenty 88 00:04:34,240 --> 00:04:37,159 Speaker 1: sixteen had a way higher standard deviation than Lamar Miller 89 00:04:37,160 --> 00:04:40,719 Speaker 1: in twenty fourteen, Breeze's coefficient of variation was point five 90 00:04:40,800 --> 00:04:43,800 Speaker 1: two while Miller's was point five to three. And that's 91 00:04:43,839 --> 00:04:46,440 Speaker 1: because Miller averaged twelve point four to eight points per 92 00:04:46,480 --> 00:04:50,279 Speaker 1: game while Breeze averaged nineteen point five two. Naturally, if 93 00:04:50,320 --> 00:04:53,200 Speaker 1: you think about how this math works, the closer the 94 00:04:53,200 --> 00:04:56,800 Speaker 1: coefficient of variation is to zero, the less volatility within 95 00:04:56,839 --> 00:05:00,200 Speaker 1: the data, because zero would imply that there's absolutely no 96 00:05:00,320 --> 00:05:03,279 Speaker 1: standard deviation. So, just to show you how predictable, or 97 00:05:03,279 --> 00:05:06,480 Speaker 1: should I say, how little variance the quarterback position has, 98 00:05:06,760 --> 00:05:09,520 Speaker 1: I divided up the position alongside the running back and 99 00:05:09,560 --> 00:05:14,280 Speaker 1: wide receiver positions and bucketed players within those positions by tiers. Yeah, 100 00:05:14,320 --> 00:05:17,080 Speaker 1: that probably doesn't make any sense. Basically, using the last 101 00:05:17,080 --> 00:05:20,159 Speaker 1: three years of weekly data which excludes week seventeen, I 102 00:05:20,200 --> 00:05:23,080 Speaker 1: grouped all QB ones by year end data together, so 103 00:05:23,200 --> 00:05:25,560 Speaker 1: QB one to QB six, and then I went down 104 00:05:25,560 --> 00:05:28,560 Speaker 1: by sixes until QB nineteen to QB twenty four. So 105 00:05:28,600 --> 00:05:30,720 Speaker 1: we're looking at QB one through QB six as one 106 00:05:30,720 --> 00:05:33,520 Speaker 1: bucket over the last three years. Then QB sevens to 107 00:05:33,600 --> 00:05:36,480 Speaker 1: QB twelve's as another bucket, then QB thirteen is to 108 00:05:36,560 --> 00:05:39,960 Speaker 1: QB eighteens, and then QB nineteen is to QB twenty fours. 109 00:05:40,000 --> 00:05:43,400 Speaker 1: In other words, I've bucketed high end QB ones, low 110 00:05:43,480 --> 00:05:48,120 Speaker 1: ENQB ones, high MQB twos, and low MQB twos. Makes sense. 111 00:05:48,400 --> 00:05:51,479 Speaker 1: Over the last three seasons, the average coefficient of variation 112 00:05:51,560 --> 00:05:53,760 Speaker 1: among the QB one to QB six group has been 113 00:05:53,800 --> 00:05:56,359 Speaker 1: point four to seven. Among the next group it's point 114 00:05:56,400 --> 00:05:58,839 Speaker 1: five to two, the next group is point five to six, 115 00:05:59,120 --> 00:06:02,160 Speaker 1: and the final group QB nineteen to QB twenty four 116 00:06:02,560 --> 00:06:05,960 Speaker 1: is point sixty four. Remember, the higher the coefficient of variation, 117 00:06:06,480 --> 00:06:10,080 Speaker 1: the more volatile the data set is, So this makes sense. 118 00:06:10,520 --> 00:06:12,919 Speaker 1: High end QB one's have the least amount of variants, 119 00:06:13,200 --> 00:06:15,880 Speaker 1: while QB nineteen to QB twenty four show the most 120 00:06:16,040 --> 00:06:19,640 Speaker 1: among quarterbacks. Doing this exercise for running backs and wide receivers, 121 00:06:19,680 --> 00:06:22,320 Speaker 1: grouping them by twelve instead of six because you're starting 122 00:06:22,320 --> 00:06:24,200 Speaker 1: at least two of them in your league instead of one, 123 00:06:24,600 --> 00:06:28,479 Speaker 1: gives us some very interesting results. Among RB one to 124 00:06:28,640 --> 00:06:31,360 Speaker 1: RB twelves over the last three seasons, the coefficient of 125 00:06:31,400 --> 00:06:34,880 Speaker 1: variation within their weekly output was point sixty four. That's 126 00:06:34,920 --> 00:06:37,039 Speaker 1: the exact same figure that we saw from the QB 127 00:06:37,120 --> 00:06:39,680 Speaker 1: nineteen to QB twenty four bucket that we just talked through. 128 00:06:39,839 --> 00:06:42,400 Speaker 1: What this means is that a random RB one over 129 00:06:42,440 --> 00:06:45,800 Speaker 1: the last three seasons from a weekly output standpoint has 130 00:06:45,839 --> 00:06:49,920 Speaker 1: shown similar variants as a low end QB two. Think 131 00:06:49,960 --> 00:06:53,520 Speaker 1: about what this means, people, Joe Flacco is basically just 132 00:06:53,600 --> 00:06:56,760 Speaker 1: as predictable as Devonte Freeman. No, I know, I get it. 133 00:06:56,760 --> 00:06:59,520 Speaker 1: It doesn't work exactly like that. David Johnson and Le'Veon 134 00:06:59,600 --> 00:07:03,640 Speaker 1: bellers certainly more predictable than Joe Flacco. But quarterbacks like 135 00:07:03,760 --> 00:07:06,719 Speaker 1: Joe Flacco are just as predictable as running backs like 136 00:07:06,839 --> 00:07:10,560 Speaker 1: Devonte Freeman. And that's pretty freaking crazy. After doing the exercise, 137 00:07:10,600 --> 00:07:13,120 Speaker 1: and this is for PPR formats, which is incredibly important. 138 00:07:13,360 --> 00:07:15,880 Speaker 1: I found that running backs start to show massive variants 139 00:07:15,920 --> 00:07:18,840 Speaker 1: after the RB two group, while wide receivers and this 140 00:07:18,880 --> 00:07:22,080 Speaker 1: is mostly again because its PPR are more consistent. A 141 00:07:22,120 --> 00:07:25,360 Speaker 1: conclusion you could probably draw from this then, is that 142 00:07:25,400 --> 00:07:28,120 Speaker 1: in a PPR league, running backs may give you an 143 00:07:28,160 --> 00:07:30,800 Speaker 1: interesting edge because if you can get an elite one, 144 00:07:30,880 --> 00:07:34,000 Speaker 1: the advantage they give you on the predictability front is 145 00:07:34,120 --> 00:07:36,920 Speaker 1: greater than what you get at wide receiver. Now, granted 146 00:07:36,960 --> 00:07:40,200 Speaker 1: that's not the case in standard formats, it's the opposite. 147 00:07:40,240 --> 00:07:43,080 Speaker 1: Another conclusion to push forth from the data is that 148 00:07:43,120 --> 00:07:46,120 Speaker 1: this is why quarterback streaming can be effective. When you 149 00:07:46,200 --> 00:07:49,160 Speaker 1: insert variables such as a good matchup, a high implied 150 00:07:49,160 --> 00:07:51,760 Speaker 1: team total, a week secondary, and so on for a 151 00:07:51,760 --> 00:07:55,160 Speaker 1: low end quarterback. But then you know that low end 152 00:07:55,200 --> 00:07:58,760 Speaker 1: quarterback is predictable, then it's easier to pinpoint when those 153 00:07:58,880 --> 00:08:03,400 Speaker 1: usable QB one weeks are coming from unusable QB twos. 154 00:08:03,720 --> 00:08:05,960 Speaker 1: This is exactly what we talk through each week on 155 00:08:06,000 --> 00:08:09,200 Speaker 1: my other podcast, Living the Stream. Over the last three years, 156 00:08:09,200 --> 00:08:12,080 Speaker 1: we've successfully streamed QB six is from a season long 157 00:08:12,080 --> 00:08:15,400 Speaker 1: standpoint by pinpointing quarterbacks owned and roughly twenty five percent 158 00:08:15,480 --> 00:08:17,920 Speaker 1: or fewer of fantasy leagues. And that's not some garbage 159 00:08:17,960 --> 00:08:21,320 Speaker 1: hindsight analysis either. That's us my podcast hosts Denny Carter 160 00:08:21,400 --> 00:08:24,680 Speaker 1: and I putting ourselves out there with picks and tracking them. 161 00:08:24,720 --> 00:08:29,080 Speaker 1: So with zero investment, you can compile a Frankenstein quarterback 162 00:08:29,080 --> 00:08:32,560 Speaker 1: who's a mid quarterback one in fantasy football. And so 163 00:08:32,880 --> 00:08:35,880 Speaker 1: much of that has to do with the predictability of 164 00:08:35,960 --> 00:08:39,640 Speaker 1: the position, but quite obviously with the coefficient of variation, 165 00:08:40,120 --> 00:08:43,360 Speaker 1: you can see trends in how particular players produce points. 166 00:08:43,480 --> 00:08:46,600 Speaker 1: Fantasy football is a weekly game, not a yearly one. 167 00:08:46,679 --> 00:08:49,760 Speaker 1: We as analysts, players and fans of the game will 168 00:08:49,800 --> 00:08:52,440 Speaker 1: often speak from the perspective of season long points, but 169 00:08:52,520 --> 00:08:55,600 Speaker 1: that tells us the end result, not the journey. The 170 00:08:55,600 --> 00:08:59,280 Speaker 1: coefficient of variation explains the journey. One of the most 171 00:08:59,280 --> 00:09:01,000 Speaker 1: common questions I I've gotten over the last couple of 172 00:09:01,000 --> 00:09:03,800 Speaker 1: months since I started this podcast is how I tier 173 00:09:03,920 --> 00:09:07,560 Speaker 1: players before a draft. My answer is always, I'll get 174 00:09:07,559 --> 00:09:10,240 Speaker 1: to it eventually on a show. I promise you this 175 00:09:10,320 --> 00:09:12,480 Speaker 1: is where I'm getting to it. Between running backs and 176 00:09:12,480 --> 00:09:16,559 Speaker 1: wide receivers. Wide receivers score points in many different ways. 177 00:09:16,920 --> 00:09:19,959 Speaker 1: Jarvis Landry over the last three years, for example, has 178 00:09:20,000 --> 00:09:22,760 Speaker 1: had a coefficient of variation that's never been above point 179 00:09:22,800 --> 00:09:25,000 Speaker 1: six to three in any given season, and in two 180 00:09:25,000 --> 00:09:27,360 Speaker 1: of his seasons it's sat at point five to three 181 00:09:27,400 --> 00:09:30,160 Speaker 1: and point five two. Just to reference the grouping method 182 00:09:30,160 --> 00:09:33,000 Speaker 1: that I walk through just a second ago, Jarvis Landry's 183 00:09:33,040 --> 00:09:36,319 Speaker 1: consistency is better than most wide receiver ones in fantasy football. 184 00:09:36,440 --> 00:09:40,320 Speaker 1: Jarvis Landry, from a consistency standpoint, is elite. But Jarvis 185 00:09:40,400 --> 00:09:43,640 Speaker 1: Landry clearly isn't an elite wide receiver in fantasy football. 186 00:09:43,679 --> 00:09:45,480 Speaker 1: The reason for that is because he's not scoring the 187 00:09:45,520 --> 00:09:48,679 Speaker 1: points that elite wide receivers score. So what does this 188 00:09:48,760 --> 00:09:51,000 Speaker 1: tell us? It tells us that Jarvis Landry is a 189 00:09:51,000 --> 00:09:55,240 Speaker 1: week to week floor play, not a seiling play. When 190 00:09:55,280 --> 00:09:57,240 Speaker 1: I'm tiering my players, I like to get a healthy 191 00:09:57,280 --> 00:10:00,520 Speaker 1: mix of players like Jarvis Landry, maybe not specifically Jarvi Sry, 192 00:10:00,559 --> 00:10:02,800 Speaker 1: but consistent floor guys where I can see a path 193 00:10:02,840 --> 00:10:06,679 Speaker 1: to more upside and weekly sealing players. I explain why 194 00:10:06,720 --> 00:10:08,600 Speaker 1: I do this on the last episode. It gives you 195 00:10:08,640 --> 00:10:11,360 Speaker 1: a nice cushion each week, all while allowing your team 196 00:10:11,400 --> 00:10:15,040 Speaker 1: to have a better ceiling if those more volatile players 197 00:10:15,120 --> 00:10:16,960 Speaker 1: end up blowing up. But I want to give the 198 00:10:16,960 --> 00:10:20,320 Speaker 1: same disclaimer that I did during the Mailbag show. Don't 199 00:10:20,360 --> 00:10:24,160 Speaker 1: overstate what this all means. Jarvis Slandry is more consistent, yes, 200 00:10:24,400 --> 00:10:26,400 Speaker 1: but he's not so consistent to the point where he'll 201 00:10:26,440 --> 00:10:29,320 Speaker 1: never post high end weekly numbers. And the opposite is 202 00:10:29,320 --> 00:10:32,080 Speaker 1: true too and PPR leagues over the last three seasons, 203 00:10:32,120 --> 00:10:35,560 Speaker 1: wide receivers who have had low coefficient of variations so 204 00:10:35,720 --> 00:10:41,679 Speaker 1: consistent wide receivers include Demarus Thomas, Antonio Brown, Richard Matthews, 205 00:10:41,920 --> 00:10:46,400 Speaker 1: Jarvis Landry, Eric Decker, Cole Beasley, and Larry Fitzgerald, just 206 00:10:46,400 --> 00:10:48,760 Speaker 1: to name a few. I don't think any of those 207 00:10:48,840 --> 00:10:51,200 Speaker 1: names are all that shocking. The thing you have to remember, though, 208 00:10:51,240 --> 00:10:53,240 Speaker 1: is that this is not telling us how good these 209 00:10:53,280 --> 00:10:56,640 Speaker 1: wide receivers are. It's telling us how consistent they are. 210 00:10:56,880 --> 00:10:58,599 Speaker 1: One of the names that was most interesting on the 211 00:10:58,640 --> 00:11:01,719 Speaker 1: list was actually Mike Wallace, But looking at Wallace's week 212 00:11:01,760 --> 00:11:04,600 Speaker 1: to week output is pretty eye opening. He hasn't hit 213 00:11:04,640 --> 00:11:06,800 Speaker 1: twenty three PPR points in a single game over the 214 00:11:06,880 --> 00:11:09,400 Speaker 1: last three years, when it's been hit three hundred and 215 00:11:09,440 --> 00:11:12,720 Speaker 1: sixty five times by a wide receiver over this span. 216 00:11:12,920 --> 00:11:15,000 Speaker 1: All the while, he's hit six or more points in 217 00:11:15,040 --> 00:11:17,120 Speaker 1: thirty eight games of the forty four game sample that 218 00:11:17,200 --> 00:11:21,240 Speaker 1: I'm analyzing, which again excludes all week seventeens, and he's 219 00:11:21,280 --> 00:11:23,480 Speaker 1: hit double digit points in over half of them. That's 220 00:11:23,480 --> 00:11:25,480 Speaker 1: why an exercise like this can be so helpful. You 221 00:11:25,600 --> 00:11:28,480 Speaker 1: end up uncovering things that you may have thought were different. Now, 222 00:11:28,520 --> 00:11:30,720 Speaker 1: an example like Wallace doesn't mean that I'll go into 223 00:11:30,720 --> 00:11:33,040 Speaker 1: a season thinking Mike Wallace is going to provide some 224 00:11:33,080 --> 00:11:35,880 Speaker 1: really great consistency on my fantasy team, but I'll certainly 225 00:11:35,920 --> 00:11:37,840 Speaker 1: be cognizant of the fact that he may not be 226 00:11:37,960 --> 00:11:40,720 Speaker 1: the boom type player that most think that he is. 227 00:11:41,080 --> 00:11:43,920 Speaker 1: So again, some guys you could bucket as consistent going 228 00:11:43,960 --> 00:11:45,760 Speaker 1: into this year, and I'll add some more players here 229 00:11:45,800 --> 00:11:49,720 Speaker 1: include Damarius Thomas, Jarvis Landry, Golden Tate is another one, 230 00:11:49,880 --> 00:11:55,000 Speaker 1: pr Garson, John Brown, Jamison Crowder. You'll find high volume 231 00:11:55,040 --> 00:11:58,600 Speaker 1: slot guys are often consistent because they're not as touchdown dependent. 232 00:11:58,760 --> 00:12:02,720 Speaker 1: On the opposite side, you have Marktavis, Bryant, AJ Green, 233 00:12:03,120 --> 00:12:07,600 Speaker 1: Sammy Watkins, Tyler Lockett, Adam Thielen Alshon, Jeffrey Marvin Jones, 234 00:12:07,840 --> 00:12:10,920 Speaker 1: and ted Gin. We know these players are deep threats, 235 00:12:11,080 --> 00:12:14,160 Speaker 1: and they're the more volatile types output wise. So let's 236 00:12:14,240 --> 00:12:16,160 Speaker 1: run through an exercise as to how I would then 237 00:12:16,320 --> 00:12:19,079 Speaker 1: draft given all of this information. Let's say I get 238 00:12:19,120 --> 00:12:21,640 Speaker 1: aj Green in the first round of my draft. He's elite. 239 00:12:21,720 --> 00:12:24,480 Speaker 1: We know that he's elite. He's going to get top production, 240 00:12:24,720 --> 00:12:27,600 Speaker 1: but he's also more volatile than an Antonio Brown type 241 00:12:27,640 --> 00:12:29,640 Speaker 1: wide receiver. If I get a running back in the 242 00:12:29,679 --> 00:12:32,560 Speaker 1: second round and then the late third round hits, I'd 243 00:12:32,640 --> 00:12:35,560 Speaker 1: end up going with a Dmerius Thomas type over say 244 00:12:35,800 --> 00:12:38,480 Speaker 1: Alshon Jeffrey. But if I were to get Alshon Jeffrey, 245 00:12:38,559 --> 00:12:41,440 Speaker 1: another up and down type player, then I'd make sure 246 00:12:41,480 --> 00:12:43,520 Speaker 1: to get a Jamison Crowder type in the fourth or 247 00:12:43,520 --> 00:12:46,400 Speaker 1: the fifth round. It's all a balancing act. This is 248 00:12:46,440 --> 00:12:49,079 Speaker 1: a big reason why I don't really agree with strictly 249 00:12:49,160 --> 00:12:52,280 Speaker 1: using season long projections to figure out rankings. You should 250 00:12:52,280 --> 00:12:55,120 Speaker 1: be reactive to how your draft is going in order 251 00:12:55,160 --> 00:12:57,559 Speaker 1: to create a roster that wins from both a ceiling 252 00:12:57,600 --> 00:13:00,960 Speaker 1: and a floor standpoint weaken and week out. This is 253 00:13:01,040 --> 00:13:04,360 Speaker 1: really an exercise that's easiest for wide receivers too, though, 254 00:13:04,400 --> 00:13:06,920 Speaker 1: a good way to think about running back consistency is 255 00:13:06,920 --> 00:13:10,240 Speaker 1: to think about receiving ability, especially in PPR formats, versus 256 00:13:10,280 --> 00:13:13,120 Speaker 1: touchdown upside. With someone like Eddie Lacy, for instance, I 257 00:13:13,120 --> 00:13:15,920 Speaker 1: wouldn't pair him with Adrian Peterson. There's a lot of 258 00:13:16,000 --> 00:13:18,520 Speaker 1: risk in doing that, since they'll probably be fairly touchdown 259 00:13:18,520 --> 00:13:20,720 Speaker 1: dependent and they may not see as much volume as 260 00:13:20,800 --> 00:13:22,880 Speaker 1: you like out of a starting running back. But the 261 00:13:22,920 --> 00:13:25,720 Speaker 1: touchdown upside is there given the offenses that they're playing in. 262 00:13:26,000 --> 00:13:29,200 Speaker 1: Pairing Lacy or Peterson in a wide receiver heavy approach 263 00:13:29,240 --> 00:13:32,040 Speaker 1: with someone like Ballall Pal that can be money. And 264 00:13:32,080 --> 00:13:34,720 Speaker 1: then make sure that you have some upside with your third, fourth, 265 00:13:34,720 --> 00:13:37,560 Speaker 1: and fifth running backs as well, so when I'm making tiers, 266 00:13:37,640 --> 00:13:40,480 Speaker 1: I'm not just looking at projections and grouping players who 267 00:13:40,520 --> 00:13:43,479 Speaker 1: are similar. On that end, I don't think it's necessary 268 00:13:43,520 --> 00:13:46,560 Speaker 1: to group Frank Gore with Adrian Peterson in a PPR league, 269 00:13:46,600 --> 00:13:49,040 Speaker 1: even if they were to hypothetically be projected for similar 270 00:13:49,040 --> 00:13:52,720 Speaker 1: season long output totals. They're two completely different running backs 271 00:13:52,880 --> 00:13:56,800 Speaker 1: and grouping that way really helps balance my roster overall 272 00:13:57,480 --> 00:14:01,280 Speaker 1: for flour and for sealing. Appreciate you guys listening to 273 00:14:01,280 --> 00:14:02,840 Speaker 1: this week's show. I hope you were able to at 274 00:14:02,880 --> 00:14:05,559 Speaker 1: least learn a little bit of new information. As always, 275 00:14:05,600 --> 00:14:07,280 Speaker 1: you can subscribe to the show by searching for the 276 00:14:07,320 --> 00:14:10,400 Speaker 1: Late Round podcast almost anywhere podcasts can be found. And 277 00:14:10,440 --> 00:14:12,160 Speaker 1: if you love the show, do me a favor and 278 00:14:12,200 --> 00:14:14,520 Speaker 1: go to iTunes and give it a rating and review. 279 00:14:14,920 --> 00:14:16,760 Speaker 1: It will really go a long way to helping the 280 00:14:16,800 --> 00:14:19,920 Speaker 1: show out. Thanks guys, I'll have a Mailbag episode out 281 00:14:19,960 --> 00:14:20,920 Speaker 1: for you later in the week.