This is the time of year I find myself obsessed with fantasy baseball. With no teams to manage, I burn off this energy on projections/$ values (now available!), building new stuff for the site (like our new team pages) and doing a lot of exploratory research/testing.

One of my favorite ‘sandboxes’ to play in is the Razzball Commenter League data. In 2014, we sponsored 84 12-team 5×5 roto leagues and we have both the draft results + end of season standing points. I have used this data set in the past to test preseason rankings (coming later this preseason) and test things like optimal hit/pitch splits in a 12-team snake draft.

I decided to test what the optimal hitter positional weight should be in our Player Rater. For our Player Rater, I compare players against the average drafted/owned player. For shallow leagues like 12-team MLB, I have been comparing a player 75% against his positional average stats and 25% against the average drafted hitter. This ratio decreases in deeper leagues to 50/50.

To date, my sole validation of this adjustment had been qualitative. I devised the following test to identify the optimal split:

  • Run the $ figures for 2013 and 2014 End of Season based on the following splits (0/100, 25/75, 50/50, 75/25, 100/0).
  • Sum the $ of all hitters greater than $0 across the 768 RCL teams (2013) and 1,008 RCL teams (2014).
    • Player worth less than $0 tend to be dropped and replaced. So this basically credits a team with $0 in those cases as it is unlikely they would have kept the anchor on their roster all year.
  • Run a correlation test on the summed hitter $ for each team against their master RCL standings points (e.g., the top team of the 1,008 gets 12 points for Runs, the next gets 11.99….last gets 1.0)

Put another way…if a player’s value should be adjusted based on his position, then it should be evident when comparing the overall value of a team with its place in the standings.

Here were the results:

Correlation % to End of Season RCL Hitting Points
Positional Weight 2014 2013
100% 66.89 63.90
75% 67.27 64.38
50% 67.56 64.83
25% 67.81 65.21
0% 67.98 65.49

The data tells a clear story – the positional factor has an negative impact on the quality of the player $ values. If I just value each player against the average drafted hitter (e.g., ignoring their position), it explains more of a team’s hitter standing points than if I factor in their position. The more I factor in the position, the worse the results.

Here is the average impact on total $ per position in 2014 for hitters valued $0+:

Position $ impact going from 75% positional weight to 0%
C -$2.8
1B +$2.0
2B -$1.1
SS -$0.8
3B -$1.1
OF $0.8
DH $1.2

While I was a bit surprised as to how small the impact is of the position weight, the fact that 1B/OF/DH benefit from no positional weight and C/2B/SS/3B are hurt makes sense.

I re-ran the above table for 2014 NL 12-team (comparing 50% vs 0%):

Position $ impact going from 75% positional factor to 0%
C -$3.3
1B -$1.4
2B -$0.8
SS -$0.1
3B -$1.0
OF +$2.5
DH N/A

The only major change is that 1B are actually helped by my position adjustment in an AL/NL-only league. But even with 2 catchers, the impact on catchers is lower than I would have guessed.

Lastly, I tested 0% vs 75% positional factor on 2015 pre-season 12 team MLB (players valued at $0+):

Position $ impact going from 75% positional factor to 0%
C -$6.4
1B +$4.6
2B -$2.7
SS -$3.2
3B +$0.8
OF +$0.7
DH +$1.8

Now that is the impact I was expecting!

I am not exactly sure why the impact is so much greater on preseason $ values versus season to date. Perhaps it is a case where the actual position averages for weaker bat positions like C/2B/SS benefit benefit more from in-season surprises than the other positions. Or because 2B/SS are less likely to be platooned than 1B/OF so they accrue more PAs. For Catcher, one driver may be that the ‘Season’ player rater reflects the ‘survivors’ (those that stayed healthy) whereas preseason factors in conservative Plate Appearance estimates across the board.

Digging into the biggest preseason ‘losers’ sans-positional adjustments, HRs are the biggest factor for MI. Troy Tulowitzki falls from $37 to $30 and HR reflect $5.3 ($14 to $8.7). Javier Baez falls a lot as well. For Catchers, it is a near across the board loss; Buster Posey lost $6 with Runs being the biggest driver at $3.

Given the results of this analysis, I removed all positional adjustments in the Pre-Season Player Rater and then looked to see if enough rostered players per position were making the $1+ threshold. Based on looking at several league formats, it became clear that Catcher was tracking just a bit low (MI is borderline). I added in a small adjustment that sets the nth catcher (where n=# of teams) at $1 and adjusts all other catchers up at the same proportion. This has a minor impact on Catcher value but nowhere near the $6 mark of the old adjustments. For AL/NL-only, I set the Teams*1.5 catcher at $1 (e.g., in 12-team NL, the 18th catcher) to avoid diverting too many fantasy $ to catcher.

While it is possible that these results are some sort of two-year fluke or that additional positional adjustments might be beneficial in some league formats, the Bayesian analysis I performed in my head determined that these test results trump any preconceived notions I had regarding positional adjustments. (And I’ve double-checked my code to make sure that the positional factors were working correctly)

I am not sure just yet what impact this learning will have on my draft strategy. Even with the prior $ adjustments, I found that I was a little more bearish than most on ‘scarcer’ hitter positions (C/2B/SS). Whenever I deviate from this, I seem to get burned (Buster Posey before Adam Jones in 3rd round of 2013 RCL, Jason Kipnis in 2nd round of 2014 Yahoo! Friends and Family). I imagine I will just be slightly more of a ‘best player regardless of position’ drafter.

 
  1. c0wfunk says:
    (link)

    here’s a real comment.. How does the fact that the difference between ie. the top catcher and the bottom catcher is a lot less than the difference between ie. the top 1B and the bottom 1Bs factor into this?

    That always seemed to me to be the real story with positional scarcity, especially now that things are starting to normalize wrt traditionally weak hitting positions like MI. Sure there are only a few good of each of these, but their potential still doesn’t get you anything like stacking up on big hitting OF and 1B at the top.

    • @c0wfunk: I don’t think that really has any factor. I think at various parts of a draft, the available hitters per position will have different distributions and that you can maximize total $ by taking a catcher in round 12 and a 3B in round 14. But the point of this article is that a 2B shouldn’t be drafted a couple rounds ahead of where you would draft him as an OF.

      With catchers, the league format is a huge driver. I think in shallow mixed leagues, the FA pool is SO huge for Catcher that it’s silly to invest money there (yet I bit last two years on Posey and Rosario…ugh). In 2 catcher formats, I can see inching a Catcher up a little bit – just nowhere near to the point where, say, 2014 Mauer becomes a first round pick (not mentioning names on who said this…wasn’t this site)…

  2. c0wfunk says:
    (link)

    “But the point of this article is that a 2B shouldn’t be drafted a couple rounds ahead of where you would draft him as an OF. ”

    I think that’s something close to like what I was trying to say :)

  3. Jerome bishop says:
    (link)

    im on your Steamer Projection page. When i click on 16 teams, it sorts all players by auction value. Which is good because although i dont play auction, i can now make a Steamer cheat sheet by manually entering every player to the appropiate position and in order by these dollar values. Im in a 20 team league so 16 teams is close enough. Am i on the right path here, or am i missing something fundamental? Thanks!!

    • @Jerome bishop: You’re on the right path! For 20-team, would be wiser to just combine 10-team AL and 10-team NL i think

  4. james says:
    (link)

    In doing a few yahoo mocks, I have noticed that it is better to just the best player…. In years past the 12th guy at any position was simply terrible…. in yahoo if you wait for SS, Xander, Baez, and Segura are all late round flyers there (or Santana goes in the 15th or so)…. if you wait for 2b you still have kendrick, walker and gyorko,…

    Simply put, the gap between #5 and #15 at so many positions has become relatively small. At one point i was all for the idea in auctions of spreading the wealth with a lot of $15-25 players, now the smart money is a few big names and then gambles.

    • @james: not sure i’m seeing major changes over the years. but i do feel people reach for MIs in early rounds. I’ve gotten Kendrick and Walker in a lot of leagues the past couple years.

  5. J-FOH says:
    (link)

    I tried doing a positional free rankings based off projections last year but I screwed it up, well I screwed up the data gathering part and my corporate analyst excel wiz baby’s mama told me to get my shizz together for 2015. Either way you are making me want to give it another go. I know even in the most well thought out strategies, I still get suckered at least once or twice by positional needs come draft day

    • @J-FOH: Your lady does your Excel? Man, that almost as emasculating as when Ms. Gamble takes out the belt sander. I mean, c’mon, there’s gotta be another way to exfoliate me, right?

      • J-FOH says:
        (link)

        @Rudy Gamble: Yes and yes. she’s works for big big corporate America as an analyst who has created tools that are used across the company. Even if I wanted to I could never get where she is, why buy the cow when you get the milk for free?

  6. GhostTownSteve says:
    (link)

    There is a difference between positional scarcity and positional adjustments to rankings, right. Positional scarcity as it is commonly called, does exist. We know intuitively that a 20 homer catcher is more valuable in draft or auction than a 20 homer outfielder. The problem with positional adjustments is that they don’t really do the thing that we need them to do. They don’t give us enough information about where the crossover point is between two players with different production levels at different positions and being considered at a certain point in a draft. It creates the illusion of an apples to apples dollar valuation so that we think we have some information about which player to select. In fact, that $ value carries little to no information necessary to make an informed draft choice.

    • Yes, position scarcity just searched more in google :)

      I agree that a static $ figure can’t identify when to choose a 2b vs a similar $ 3b because there are more 3b priced similarly.

      That is where ADP helps in mapping out a strategy.

    • Bull in a Chinese Restaurant says:
      (link)

      @GhostTownSteve: they are way more prevalent in football, but there’s those VBD (value at position) draft software programs out there. You could do these calculations yourself, but not in 90 seconds, for every round live.

    • dingus says:
      (link)

      @GhostTownSteve:

      Fantistics used to have an in-draft tool that would recalculate VAM (value above mean) throughout the draft. This would tell you the point where Kolten Wong’s 15/15 potential is worth more (versus remaining 2B) than Player X in the OF. Even if he started the draft with a lower value, as people stress ‘position scarcity’ by over-drafting 2B…the value gets deeper in OF right?

      So Wong’s increased VAM doesn’t necessarily mean he’s the right pick (bad choices by others may have inflated his value) but it does provide in-draft visibility to remaining players and their relative value.

      I think it could be set up with or without position adjustment. Been a decade since last used it…

  7. TheTinDoor

    TheTinDoor says:
    (link)

    I added in a small adjustment that sets the nth catcher (where n=# of teams) at $1 and adjusts all other catchers up at the same proportion.

    I don’t quite understand – what was the positional adjustment like prior? The system above (making sure that all roster spots are at least $1) is kind of the default, right?

    In a two-catcher league, the 24th-ranked catcher can be pretty far in the ‘negative’, causing the move-to-$1 adjustment to be quite big for the other catchers. It makes math sense, but doesn’t pass my sniff test

    • TheTinDoor

      TheTinDoor says:
      (link)

      @TheTinDoor: Never mind, I see that you’ve been comparing within the position AND to the overall pool.

    • Yes, that is generally standard in methodogies that compare players to replacement value. Mine uses average value and just never felt that adjustment was necessary. I think some overdo those adjustments though. I find it suspicious when i see too many $1 values – clearly some late picks are better than others.

    • Blue says:
      (link)

      @TheTinDoor: And it shouldn’t pass your sniff test either. Fangraphs, for example, adds $6 to all catchers to try and meet replacement level. But if you go and actually look at how that plays out in a real draft you end up overpaying massively for a marginal gain.

  8. The Committee that agrees with Rudy says:
    (link)

    I concur completely on position scarcity taking a back seat and have a couple root cause theories to put forth.
    1. The overall drop in offense has made getting quality offensive players from any position important because they leave the board very quickly.
    2. Hand-in-hand with that I have a suspicion that over the last 2-3 years the positive end of the “bell curve of hitters” is now shorter and thinnerwhile the middle cluster around “major-league average” has fattened and the negative slope is larger and less steep than the positive slope.

    You can almost compare the state of today’s offense to a man passing middle age. Belly getting bigger, butt getting bigger and dick shrinking. Oh the inhumanity of it all…..

    • Yescheese says:
      (link)

      @The Committee that agrees with Rudy: actually I think this is what the data is really suggesting … What we have is not position scarcity, but rather, power scarcity.

      The only positions that consistently deliver value in that scarce world are 1B and OF, and OF is constrained by artificial depth (3-5 OF positions per team).

      Were the data available, I’m confident you’d see a greater correlation to OF1 players vs OF4/5 types.

      Offense is scarce everywhere. Guys that can deliver on that variance are gold. Where does 2014 Anthony Rendon come out in this analysis?

      • Well rendon outperformed expectations – no position factor is that big (a league requiring two hitters with 2B/3B eligibility?)

        The same happened in the inverse with Kipnis.

        But those are the extremes vs the norms

    • I think there might be something to it. Feels like no position is deep but there are still decent late round picks at every position.

  9. Hoosier Daddy says:
    (link)

    Rudy-

    I’m in a 2 catcher league 12 team roto with 5OF and a corner and middle man with limited trades. With most trades used for picking up pitching, I’d prefer not to waste trades on catchers. I typically feel I need one good catcher and can punt the second. If I was to target one good catcher, …assuming he’s ranked ~~100 here, how much would you bump him up in such a league? 20 slots? one round?

    • That is a solid strategy. That is my norm for 2 C leagues. one decent, one real late. BUT, i would try to pay no premium for 1st C. Best case, you value the catcher slightly more than others and get hik where u want. worst case, you know people are waiting on C2 and may be able to get a nice discount on the top C2s as your C1.

  10. TheTinDoor

    TheTinDoor says:
    (link)

    Is there any chance you can share access to that “Sandbox” of RCL data? I’m interested in investigating a few draft strategy questions (first up would be should you “pay for saves” or not), but sample size is the major hurdle.

    • It might be easier if I let everyone submit their theories and I run them.

      • TheTinDoor

        TheTinDoor says:
        (link)

        @Rudy Gamble: Consider this one vote to answer the saves question. I assume there would be some correlation between $ spent on closers and final standings points in the Saves category, but is it a strong one? More importantly, what’s the impact on overall standings results?

        I tend toward the Larry Schecter opinion (just get value in whatever form it appears), but would be interested to investigate

        • Okay, let me start up a new post this week to gather up the ideas.

  11. Not surprising to me in the current MLB Player pool. Wonder how those numbers looked in the days of Piazza and Pudge, Arod, Nomar and Tejada?

    I also find that people add value to all the players at a “thin” position. The only added value (if any) should be to the scarce elite options. I’m not gonna value Jhonny Peralta higher just because shortstop is thin… I’d certainly give Tulo a boost if I trusted his health.

    • I think Piazza’s stats were just elite – regardless of any positional adjustment. I agree that i’d be a little more likely to go a $1-$2 higher for an elite talent at weaker position but this exercise might’ve snapped me out of that

  12. Holden says:
    (link)

    I wonder if you have a mistake in your process, it’s hard to tell from what you said, but if you considered only the players > 0$ then the positionally adjusted numbers will include more lower quality players overall, and the non adjusted numbers will cut out a ton of lower producing guys.

    • That is possible to a degree. The position adjustments also distort the true value of players in CI/MI/UTIL.

      I will check the average players per team and $ per team to see what the diffs are

  13. What does this mean for a lay person? How would an idiot use this? I’m not an idiot, but I’m lost reading this.

    • @Campbell Sisson: Basically, the analysis showed no proof that you should adjust a player’s fantasy baseball value based on his position. So the most actionable aspect of it is to avoid overpaying for hitters at weaker positions.

  14. dingus says:
    (link)

    Isn’t it more about SEPARATION than anything else? Will Abreu/Rizzo have more raw stat production then Tulo/Cano types? Certainly. But will they give you more of an advantage vs the opportunity cost?

    Say Rizzo vs Hosmer/LaRoche/Belt tier is +10-15hr, +20-30r&r, wash in average. Second round you’re facing Rizzo vs Cano (or Desmond, etc). The question should be what expectations are for Cano vs a similar tiered 2B. If the gap is greater than Rizzo vs 1B, go with Cano. NOT because his position is scarce and also NOT because he’s the best stats available (which he wouldn’t be).

    THIS is where sleepers matter! If you believe Wong is a breakout value, then cross Cano off your fricken draft sheet. Pencil in Wong a couple rounds early and you’ve ‘kept up’ with every team that drafted a 2B before you.

    Tulo vs lesser SS might give more separation than even Abreu vs lesser 1B. That’s where ‘scarcity’ matters IMO. Soon as TT’s gone though, does the same logic automatically apply to Ian Desmond vs Rizzo in round 2? Maybe. Maybenot.

    Which player gives me the most separation? That’s the primary question I ask myself for each pick.

    • Robdouth says:
      (link)

      @dingus:

      You make a great point, but isn’t it about targeting for a specific number of stats. It’s why the draft cheat sheet is designed the way it is. You are trying to get to a specific number of counting stats/avg/HR/etc if you are roto, and even though that’s not the way H2H works, it is still a good barometer of how you should fall in that category H2H during a given week.

      • Yes – getting the right mix of stats (or using the category $) is the key. I had felt that adjusting at position level would help there (so tulo HR more valuable, hamilton’s lack of power for OF that much worse) but cant find proof it does

      • dingus says:
        (link)

        @Robdouth:

        “You are trying to get to a specific number of counting stats/avg/HR/etc”

        What?! I’m trying to win the league, not draft a perfectly balanced roster.
        1. You know a good portion of drafted players won’t perform to expectation
        2. Injuries will certainly mess with a drafted roster

        If all someone did was take the MAX-separation player, who cares what sort of counting stats are projected??! Seriously. If your league ‘gives’ you value at SP, take it. If in the OF, take it.

        Best-value drafting should lead to imbalanced stat projections. That’s what trading is for, and with value stored up a good manager should be able to evaluate standings and go get what’s needed in-season, no?

        Kills me when people pass up a 3 or 4 category bat, or a quality pitcher, to grab Player X for his ‘average and steals.’ Waaaay too much focus is put on designing a perfect team out of the draft. Let that go!

        • Having category targets isn’t a bad thing but, yes, slavishly trying to balance ur squad throughout draft can be short-sighted. I do consciously keep speed and AVG into account to avoid getting too far behind in those cats.

    • Blue says:
      (link)

      @dingus: Nope. Separation doesn’t really matter. Accumulating counting stats is what matters.

  15. Robdouth says:
    (link)

    Is there any way to tell what the chances of winning are if you draft a pitcher in the first/second round. We have all heard, and I think most agree that Kershaw in the top 3 doesn’t make sense, but is there a way to data mine the chances of winning in a full redraft league (or just no keepers league) where Kershaw was taken top 5 and see the chances of that team winning vs. just a normal statistical distribution. I’m sure some guys who drafted Kershaw won because some hitters outperformed their draft spot, but I would think it’s safe to think they had a lower % chance of winning their league than others.

  16. mrrr says:
    (link)

    Rudy – are you planning to host a Razzball NFBC challenge again? I’m

    • I think so – need to check in with my contact there

  17. Jerome bishop says:
    (link)

    Im going to import players from the Steamer projections into a cheat sheet format. My league is 20 teams so i should use the AL/NL 10 team format. However, its more difficult since i have to cross check $values when combining two separate databases into one. For a snake draft league, how important is it to use the 20 team auction values as opposed to the 16 team values? It seems to me that the ranking of players in a snake draft is more important than the $ values. Do you urge me NOT to take the 16 team short cut when importing players to Excel? Thank you, Rudy.

Comments are closed.