In my last post, The Legend of the Bid Button, and the first in this series, I examined and explained my strategy for approaching auction drafts. The key stat that I rely upon is “points per dollar” (PPD). How many fantasy points a player is expected to get me for each auction dollar I spend on him. If you have not read that post, please take ten minutes and do so. For those that have already done so and are back for round two, welcome back. While we have ten minutes to kill as we wait for everyone else to catch up, here is a link to a sub-five minute speed run of Super Mario Brothers. Feel free to watch it twice. I know these guys use tools to help accomplish this, but as someone who played this game back when it first came out, this is still pretty awesome.

Ok, now let’s get back to our regularly scheduled programming. I’d like to take a look back at last season’s PPD heroes and zeros…

For the sanity of this content I am going to primarily be targeting players that had a minimum cost of $5. The reason for this is that otherwise, one of the many $1 and $2 players would certainly skew these results. However, I will list any sub-five dollar player of significance.

Quarterback

Player Team FPTS Cost PPD
Matt Ryan ATL 282.26 $6.4 44.103
Colin Kaepernick SF 250.66 $9 27.851
Tom Brady NE 278.06 $11.2 24.827
Andrew Luck IND 351.74 $17.7 19.872
Cam Newton CAR 246.98 $14.8 16.688
Matthew Stafford DET 249.58 $19.9 12.542
Nick Foles PHI 119.32 $10.8 11.048
Robert Griffin III WAS 87.36 $9.5 9.196
Aaron Rodgers GB 352.14 $47.1 7.476
Drew Brees NO 302.88 $43.3 6.995
Peyton Manning DEN 308.68 $54.1 5.706

Peyton Manning was the most expensive quarterback at $54.1 and he yielded the lowest PPD of all quarterbacks that cost at least $5. The QB with the overall worst PPD was Johnny Manziel who put up 11.9 points and cost $2.3 for a PPD of 5.174. Johnny on the spot!

So who was the best bargain last year. Reading the above table is not as easy as sorting by PPD and picking the guy at the top. At least not in all cases. Heck, there isn’t even a definite right answer. Based on the numbers I see two possible answers. The first is Matt Ryan. For just $7 (I am rounding up since if his average cost was $6.4, you’d need a bid of $7 to win him), he got you 282 points. That’s a PPD of 40.285. Only four QBs scored more points than Ryan, and they all cost more. A lot more in most cases. Rodgers, Brees and Manning cost significantly more. Andrew Luck however only cost about $18 and he ended up with the second most points amongst quarterbacks with 351.74 points. Looking at the numbers, you “might” have been able to convince me to take Luck over Ryan. Not likely, but I can see the argument. For $11 extra dollars you could have had nearly the top QB in fantasy football and 70 additional points. I still think I would go with Ryan and use that $11 elsewhere.

The quarterback with the highest PPD regardless of cost was Eli Manning. At $2.1, Eli scored 269.5 points, boasting a PPD of 128.33. Let’s take a look at the other QBs that scored about 270 points and see how much they cost as compared to Eli.

Tom Brady, 278 points, $11.2 (24.827 PPD)
Ryan Tannehill, 278 points, $2.9 (96.172 PPD)
Tony Romo, 266 points, $4.7 (56.66 PPD)
Philip Rivers, 265 points, $3.8 (69.905 PPD)
Joe Flacco, 262 points, $2.8 (93.729 PPD)

I feel bad for the owners that spent $11.2 on Brady. I bet you they missed out on Le’Veon Bell by only $5. What a shame!

Heroes: Matt Ryan and Andrew Luck
Zero: Peyton Manning

Running Back

Player Team FPTS Cost PPD
Joique Bell DET 198.2 $8.9 22.27
Shane Vereen NE 165.8 $8.6 19.279
Rashad Jennings NYG 138.5 $11.3 12.257
Bishop Sankey TEN 96.2 $8.1 11.877
Chris Johnson NYJ 115.4 $10.5 10.99
Andre Ellington ARI 177.5 $16.4 10.823
Le’Veon Bell PIT 370.5 $34.4 10.77
Trent Richardson IND 117.8 $11.6 10.155
Frank Gore SF 158.7 $17 9.335
DeMarco Murray DAL 351.1 $39.1 8.98
Giovani Bernard CIN 187.9 $27.8 6.759
Alfred Morris WAS 187.9 $29.6 6.348
Matt Forte CHI 342.6 $54.6 6.275
Arian Foster HOU 269.3 $43 6.263
Marshawn Lynch SEA 302.3 $52.3 5.78
Eddie Lacy GB 272.6 $50.5 5.398
Reggie Bush DET 107 $24.5 4.367
Jamaal Charles KC 250.4 $58.3 4.295
LeSean McCoy PHI 200.5 $60.5 3.312

Continuing with an earlier theme, notice that the most expensive running back, LeSean McCoy, has the lowest PPD. There seems to be a pattern developing here. So who stands out on the plus side? It’s got to be Le’Veon Bell. Even at $34 his PPD comes in at 10.77 due to his 370 points. Stud running backs usually cost about $50, so getting Bell at $34 was a steal. After Bell comes Bell. Joique Bell. With a PPD of over 19, he cost only $9. While not an RB1, he was certainly worth his price tag.

How about the cheaper running backs? Mr. Anderson, welcome back! With a PPD of 192.091, C.J. Anderson was perhaps the most valuable player in fantasy football last season. This, however, is a prime example of a $1 player with an exaggerated PPD. But is it really exaggerated? I say no. His cost was only a dollar. Why should a bargain be penalized for being a bargain. He shouldn’t, but it also doesn’t make sense to compare these players with those with legitimate price tags. The other issue is that most of these $1 players weren’t actually drafted in most auctions. Instead they were waiver wire pickups. That means that their value was not predictable prior to last year’s auction.

Here are a few others:

Jeremy Hill, 211 points, $1.4 (150.643 PPD)
Mark Ingram, 192 points, $1.6 (119.938 PPD)
Lamar Miller, 223 points, $2.8 (79.786 PPD)
Fred Jackson, 187 points, $3.9 (47.872 PPD)
Darren Sproles, 146 points, $4.2 (34.667 PPD)

Heroes: Lamar Miller and Justin Forsett (245 points, undrafted)
Zero: LeSean McCoy

Wide Receiver

Player Team FPTS Cost PPD
Golden Tate DET 259.1 6.4 40.484
Eric Decker NYJ 200.2 5.3 37.774
Jeremy Maclin PHI 276.8 8.2 33.756
Emmanuel Sanders DEN 299.8 8.9 33.685
Sammy Watkins BUF 198 6.4 30.938
T.Y. Hilton IND 258.5 8.6 30.058
Marques Colston NO 177.2 6.9 25.681
Julian Edelman NE 222.6 8.7 25.586
Torrey Smith BAL 191.7 9.7 19.763
Michael Floyd ARI 165.3 8.9 18.573
DeSean Jackson WAS 209.6 14.6 14.356
Antonio Brown PIT 374.1 26.1 14.333
Roddy White ATL 212.1 15.1 14.046
Michael Crabtree SF 162.2 11.8 13.746
Randall Cobb GB 291.4 23.6 12.347
Jordy Nelson GB 327.9 30.7 10.681
Wes Welker DEN 107.4 10.1 10.634
Keenan Allen SD 175.3 17.1 10.251
Percy Harvin SEA 129.5 13.2 9.811
Pierre Garcon WAS 161.2 16.5 9.77
Alshon Jeffery CHI 261.6 27.8 9.41
Vincent Jackson TB 180.2 20.1 8.965
Andre Johnson HOU 190.6 21.7 8.783
Julio Jones ATL 297.4 34.8 8.546
Cordarrelle Patterson MIN 93.1 11.2 8.313
Demaryius Thomas DEN 338.9 42.9 7.9
Dez Bryant DAL 316 40.1 7.88
Larry Fitzgerald ARI 151.4 20 7.57
A.J. Green CIN 205.3 41 5.007
Brandon Marshall CHI 179.1 36.4 4.92
Calvin Johnson DET 226.7 54.7 4.144

Surprise, surprise! Any guesses which wide receiver had the lowest PPD? If your answer is the one that cost the most, you’d be correct. MegaTron was MegaExpensive at just under $55, and with 227 points, his 4.144 PPD is pretty abysmal. The players that stand out to me are Antonio Brown, Golden Tate, Jeremy Maclin, Emmanuel Sanders and Randall Cobb.

And the cheap seats?

Brandon LaFell, 210 points, $1.4 (150.429 PPD)
Andrew Hawkins, 159 points, $1.2 (132.417 PPD)
Rueben Randle, 183 points, $1.4 (130.571 PPD)
Doug Baldwin, 167 points, $1.3 (128.692 PPD)
James Jones, 177 points, $1.4 (124 PPD)
Kenny Stills, 174 points, $1.5 (115.933 PPD)
Steve Smith Sr, 219 points, $2, (109.5 PPD)
Mike Evans, 245 points, $2.3 (106.565 PPD)
Jordan Matthews, 202 points, $2 (101.1 PPD)
Kelvin Benjamin, 226 points, $2.8 (80.643 PPD)
Anquan Boldin, 220 points, $3.3 (66.545 PPD)

Heroes: Antonio Brown, Emmanuel Sanders and Odell Beckham Jr. (295 points, undrafted)
Zero: Calvin Johnson

Tight End

Player Team FPTS Cost PPD
Delanie Walker TEN 176 1.1 160
Antonio Gates SD 223.1 1.9 117.421
Martellus Bennett CHI 217.6 1.9 114.526
Coby Fleener IND 176.4 1.6 110.25
Charles Clay MIA 136.5 1.4 97.5
Heath Miller PIT 158.1 1.8 87.833
Zach Ertz PHI 144.2 2.2 65.545
Greg Olsen CAR 220.8 3.9 56.615
Jordan Reed WAS 94.5 2.3 41.087
Eric Ebron DET 55.8 1.8 31
Ladarius Green SD 41.6 1.9 21.895
Jason Witten DAL 164.3 8.6 19.105
Kyle Rudolph MIN 59.1 3.1 19.065
Jordan Cameron CLE 78.4 6.4 12.25
Rob Gronkowski NE 266.4 22.5 11.84
Dennis Pitta BAL 28.5 3.3 8.636
Julius Thomas DEN 163.9 26.5 6.185
Jimmy Graham NO 229.9 46.5 4.944
Vernon Davis SF 62.9 13.7 4.591

Leave it to Vernon Davis to ruin the streak. Could that dude have been more disappointing last year? That was a rhetorical question. I’ve included all tight ends in this list since it’s a much shallower position and many teams might only take one.

For $2 Antonio Gates got you 223 points. For $4 Greg Olsen got you 220 points. But if you felt like “making it rain” on auction day you could have spent $46 for Jimmy Graham and gotten 230 points! You do the math.

Hero: Antonio Gates and Greg Olsen
Zero: Vernon Davis and Jimmy Graham

They say “hindsight is 20/20”. While I agree with that statement, I’m still trying to figure out who “they” are. Let’s take a look at some players that were selected in auctions and find another player that had the some production for less money. I realize this was impossible to predict. The purpose here is to highlight how PPD can be used to compare players.

LeSean McCoy (200 points, $60.5)
Joique Bell (198 points, $8.9)

Holy waste of 51 dollars Batman!

Alfred Morris (188 points, $29.6)
Giovani Bernard (188 points, $27.8)
Fred Jackson (187 points, $3.9)

If you took Jackson here, you could have called your team Right Said Fred!

Frank Gore (159 points, $17)
Shane Vereen (166 points, $8.6)

This one isn’t too gory, but I still suck. You would have been much better off with Vereen.

Julio Jones (298 points, $34.8)
Emmanuel Sanders (300 points, $8.9)

Sanders production makes me feel like I did when I was a kid and stumbled across Emmanuelle on cable tv.

Alshon Jeffery (262 points, $27.8)
Jeremy Maclin (277 points, $8.2)

Nearly a $20 difference. Holy mac-a-lin!

Calvin Johnson (227 points, $54.7)
Kelvin Benjamin (226 points, $2.8)

MegaDisappointing! Unless you had Benjamin.

Roddy White (212 points, $15.1)
Mike Wallace (213 points, $5)

I bet you wish you had that $10 back so you could have gotten Antonio Brown.

Andre Johnson (191 points, $21.7)
Torrey Smith (192 points, $9.7)

By now I’m sure you get the point.

One last thing I’d like to leave you with while we’re on the topic of hindsight. Here’s the team that I wish I had ended the auction with. And yes, I’ve decided to take Andrew Luck. This team is unbeatable. I’d say the only thing you’d have to worry about is who to drop when Luck is on a bye, but with this team you could probably just take a zero at QB that week! Team name, Antonio’s Lucky Bells

POS Player Cost Points
QB Andrew Luck $18 352
RB Le’Veon Bell $35 371
RB DeMarco Murray $40 351
WR Emmanuel Sanders $9 300
WR Antonio Brown $27 374
TE Antonio Gates $2 223
FL Jordy Nelson $31 328
K Who Cares $1 1
DST Doesn’t Matter $1 1
B Jeremy Maclin $9 277
B Golden Tate $7 259
B Joique Bell $8 198
B Lamar Miller $3 223
B Jeremy Hill $2 211
B Mike Evans $3 245
B Greg Olsen $4 221
$200

 

 

  1. RotoLance

    RotoLance says:
    (link)

    Another good post in a really useful series but I want to debate you on this point:

    “For $11 extra dollars you could have had nearly the top QB in fantasy football and 70 additional points. I still think I would go with Ryan and use that $11 elsewhere.”

    For this to be the correct strategy it means that you’ll need to find another position where you can take a solid starter and spending only $11 more turn that player into a stud by adding 70 more points to the solid starter. And I doubt that’s going to happen using projected points and dollar values.

    So yeah, given you changed your mind at the end… you’ve got to know there was an error in or original line of thought on Ryan +$11 vs Luck. Honestly as soon as I read that I thought that was a very bad decision. Ryan is too close to replacement level.

    That’s why maybe you should come up with some sort of PPD over replacement metric and see if that helps make the decisions when players are not of similar performance. So regarding Matt Ryan again, part of the reason you need to think about replacement level is because although “Only four QBs scored more points than Ryan” of those $5 or greater, Roethlisberger and Wilson outscored him and Tannehill was only 4 points off. So really Matt Ryan was more or less middle of the pack as far as a starting fantasy QB. That’s ultimately why you spend that $11 more for Luck (in hindsight). And I think when putting together that team at the end it should be apparent that Ryan isn’t really that big of a difference maker, a useful piece that’s for sure, but not a difference maker like Luck was for $18.

    • malamoney

      malamoney says:
      (link)

      @RotoLance: Hey man. Thanks. This is a fantastic comment. I am definitely going to give this some deeper analysis. And yes, at the end I decided to go with Luck. But that was with hindsight in the rear view. The fact that Luck cost only $18 and ended up being nearly the top QB is not something that is easy to predict. Most were expecting the top QB to be Peyton or Rodgers. With even Brees outperforming Luck. That’s why they were all so expensive. Should you pay $11 more for a stud over a “replacement” player? Absolutely. But like I said, there way no way to know that $18 Luck was going to provide that. In order to get that at auction time, you would have needed to spend the big bucks on Peyton or Rodgers.

      The purpose of this post and the concept of PPD is to give you an extra data point to help you make decisions are auction time. Once the auction is over, PPD is pointless. You can look back and see how a player’s PPD actually ended up, but at this point, it is of no help to your cause.

      Thanks again for taking the time to provide your thoughtful analysis and please keep it coming…

      • RotoLance

        RotoLance says:
        (link)

        @malamoney: Well, I’m glad you appreciated it. As long as your posts continue to be thought provoking I’ll keep it coming (well, ok.. time permitting).

        I want to point out for anyone else reading this, that technically, I never called Matt Ryan a replacement level player and I don’t think that’s what you meant either. He’s a solid starter.

        So that brings to mind something related… whether you pay $11 more for 70 points above “replacement level” or you pay $11 more for 70 points over “middle of the road starter” in one sense doesn’t matter because they will both be well worth it. But in another sense I think it’s probably more meaningful to your team to pay $11 more over the middle of the road starter. That’s something a good PPD over replacement type metric might help us figure out.

        My home league uses an auction format so I’m pretty experienced with auctions. The whole reason you pay up those big dollars for stars is related to points over replacement, so while PPD is useful when players are about the same I think it could lead folks down the wrong path. That’s why I appreciated, as I pointed out already, how you really made it a point of emphasis in that first post that you can only use PPD for players of similar projected performance level. It’s super useful for that, but very misleading otherwise, at least to the uninitiated.

        I could very well write my own post on auctions to complement yours but I don’t think I’ll have time. There’s some things I’ve thought about over the years, auction strategy wise, that I think could be pretty useful to people and it touches on different aspects than what you’ve covered so far.

        • malamoney

          malamoney says:
          (link)

          @RotoLance: On the baseball side of Razzball I generated some spreadsheets with respect to replacement level players and z-scores. I will see what I can do here on the football side of things.

          Thanks again…

  2. TheStu says:
    (link)

    I’ve really enjoyed this series. When I look up auction values (projected and average) they seem to be all over the map. Some sites have the top guys going for 50-60 and others have them at 40-50. How/where do you collect your auction values and projected points? I play on Yahoo mostly, but I don’t want to solely rely on just their data.

    • malamoney

      malamoney says:
      (link)

      @TheStu: Thanks. You can use projected auction values to determine PPD. You can, but it would be less than desirable. Nearly all fantasy football site, Razzball included, publishes a cheatsheet of how much they “think” a player is worth. This is a somewhat helpful guide, but we can do better.

      What I do is I download the results of thousands of completed online auctions and use those to determine how much a player is actually going for in real auctions. It’s one thing for a website to say “Andrew Luck is worth $29”. But what I really care about is the fact that Andrew Luck is actually going for $36 on average across thousands of actual auctions. This data is much more useful to me. It gives me a much closer idea of how much I am going to have to spend to get that player. Now that doesn’t mean he won’t go for more or less, but it’s the best baseline we’ve got.

      I am in the process of putting together my own projections. I will then turn those projections into points and use current auction values to determine PPD. I should have this post out by next week…

      • Montezuma's Revenge... right now says:
        (link)

        @malamoney: this is effectively the ADP auction price rather than just taking some site’s expected/projected price. much better.

Comments are closed.