Baseball Professor header image
≡ Menu

How Billy Hamilton Is Changing Fantasy Rankings Systems

How many times will we see THIS in 2014?

How many times will we see THIS in 2014?

Work has begun on our 2014 fantasy projections and we’ve already hit a snag in the form of a 160-pound outfielder for the Cincinnati Reds.

Put bluntly, Billy Hamilton is going to break projection systems across the web.

Since the Sabermetric Revolution began a little over a decade ago, we haven’t seen a player with Hamilton’s stolen base potential. In fantasy leagues, valuing Hamilton’s contribution in steals is going to prove difficult. It’s not a bold prediction to say that by Opening Day, 2014, Tristan Cockcroft will address this issue.

You see, there’s a non-zero chance that Hamilton gets Cincinnati’s center field job and approaches or exceeds 100 steals. In fact, non-zero probably doesn’t accurately reflect how likely this scenario is. I wouldn’t say it’s probable, but it’s definitely more than plausible.

And if you plug 100 steals into most conventional projection systems, Hamilton will come out as a top-five player. (We tried it. He did.)

Look, those 100 steals are valuable. You can flat-out win that category from wire-to-wire solely because of Hamilton, but don’t forget that he’s probably not the only guy you’re going to have who steals bases. At what point have you gone beyond fortifying the category?

I’d argue that stolen base benchmarks 70, 80, and 90 are less valuable than numbers 30, 40, and 50 — at that point you’re just needlessly piling it on — so for me the question is, “How much does the value of a steal change as a player keeps accruing them?”

Essentially, I’m talking about the Law of Diminishing Returns as it pertains to fantasy stats. I sat down and attempted to answer this question.

Video after the jump!

facebooktwittergoogle_plusredditpinterestlinkedintumblrmail

About the author: Bryan is the co-founder of Baseball Professor and works as a consultant specializing in operational metrics and efficiency analysis. When he’s not working, blogging, or tending to basic human needs, he enjoys pondering the vastness of the universe, rewatching episodes of Breaking Bad, and avoiding snakes. (@BaseballProf)

Comments on this entry are closed.

  • Bruce

    The real value of a player who can win a category all by himself is that it allows an owner to focus on the other categories, weighting the rest of the players selected more toward power and average than speed, or bumping a pitcher or two up instead of reaching for a guy with medium legs. Add to that the fact that if Hamilton has anything more than a poor to mediocre OBP, he’s going to score a LOT of runs (even more so because he’s playing for a good offense), and his value really is very high. Perhaps not top-5 in the same way a Mike Trout or other five-category players are, but honestly very high (remember Vince Coleman in 1986 and 1987 — that was when I started to play). Systems might have to be changed to accommodate him, but downgrading him too much is a mistake, too.

    • http://www.baseballprof.com/ Bryan Curley

      Hey Bruce. I would agree to an extent. You can never rely on one player for all your steals because you’re screwed if he gets hurt, but it does allow you some more flexibility for sure. Thankfully even this kind of analysis knocked him down one spot Inn the rankings so it wasn’t a massive over compensation. Thanks for reading!

  • http://fantasybaseballcrackerjacks.com/ Clave

    Love it. I wrote a piece once where I looked at Billy Hamilton if he existed solely as a pinch runner. It’s remarkable the value he could add if he never even lifts a bat.

    This is a good add for your rankings and I’m looking forward to see what you guys crank out for 2014.

    I will say that I’m with Bruce. You could lock up steals with Hamilton, then couple him with a cheap Brandon Moss and a mid-priced Matt Carpenter. Suddenly you’re winning in steals, Moss adds power, you’re crushing in runs, and Carpenter helps pull the average back up to respectability.

    Substitute whatever players you like, as I simply shared this example to illustrate my excitement with how a player like Hamilton adds incredible flexibility when constructing a roster.

    • http://www.baseballprof.com/ Bryan Curley

      Hey Clave,

      I’m 100% on board with you and Bruce. Like I said below I think it’s exceedingly risky to put almost all your SB potential in one player (because if he gets hurt you’re sitting in the bottom 25-30% of the league), but Hamilton should produce enough with his legs and defense to stay in the majors once he gets a full-time job. No Dee Gordon concerns here from me!

  • Fred

    This is a great article that really describes the value of Hamilton in roto. Having Hamilton on your team presents you with two scenarios, either count on his steals as a given and draft the low average bombers like Trumbo and Bruce or be safe and ignore his excess SB’s, continue drafting 5 category players and work the trades over the season, getting rid of your excess steals and get the lower value boppers like Moss. Either way you have great options on draft day.

    • http://www.baseballprof.com/ Bryan Curley

      He’s a very unique player. I’m curious if he bats leadoff with, say, a .330 OBP and 70+ SB, how many runs scored does that equate to? He’ll almost always be on 2nd, and Phillips/Votto/Bruce are great run producers. Could we be looking at a 130 run season, too?

      • Paul D’Amore

        Hi Brian,
        I do something similar in my z score where I cap the stdev on sb’s to account for the fact steals lose relative value as you slide up the scale. My question is more holistic. In your analysis you say that Hamilton’s 80th steal is only 86% as valuable as the first, but his score only drops enough to have him lose one spot in the hypothetical rankings? I am not saying it is wrong but doesn’t pass my initial sniff test. To me if his one main asset loses it’s effectiveness by 15% then the impact to the sb portion of his score should be impacted more than it appears to be. And conversely if home runs are as effective in gaining relative value, and Hamilton hits three all year that having someone in the lineup like Hamilton will negatively impact home runs. To me even at 80 sb I find it hard to believe that Hamilton is anything close to a top 10 player.

        Did you adjust down the impact his noodle bat will have on the hr and RBI portion of his score? As your graphs show, hr add value as they move up, so would assume sliding down the scale would bring his score down further.

        LOVE this analysis and welcome the feedback. Keep up the great work.

        Paul

        • http://www.baseballprof.com/ Bryan Curley

          Hey Paul. Thanks a lot for the great comment. Here are my thoughts on some of your points:

          1) The 15% decrease is only for steal number 100, not for all the steals that precede it. The first 60 steals are worth about the same, and then each one drops slightly after that. At 100 steals, this analysis equates them to 91 steals in a system that weighs all steals the same. Make sense?

          2) Yes, his HR and RBI totals were projected in the stat lines I showed in the video (4 HR, mid-30s RBI in one scenario, 2 HR, mid-20s RBI in the other) and that was factored into the rankings you saw at the end of the video.

          3) His low HR total doesn’t bring his score down. Instead, it increases everyone else’s scores. We saw this with the Trumbo and Alvarez examples in the rankings.

          4) It should also be noted that if Hamilton approaches 80-100 SB, he’ll probably be around 80-100 runs scored too (minimum).

          5) The one thing this analysis does not capture (yet!) is how you’d construct your team planning on Hamilton getting more SB (therefore targeting more high-HR players). This would actually INCREASE Hamilton’s value, though, so it’s not something we’d address if we think his value is already too high even with this analysis. Perhaps I can include some kind of “readily available replacement level productivity” correction, but I’m not 100% sure how I would do this yet.

          6) The numbers in the video are correct. I checked them many, many times. In the end, a player that scores 100 runs and steals 80 bases is going to be valuable in any rankings system.

          7) I’m not fully sure what you mean by capping the stdev on SBs in your system. I attempted to do this by separating players who stole fewer than 2 SB/600PA, therefore only looking at players who actually steal bases. Perhaps I need to raise my minimum SB requirement to 5? I honestly haven’t looked at that yet.

          Again, thanks for the comment! Sorry if this was overly wordy!

  • WisBrave

    Great article, after reading this I’m on the fence about keeping B.Hamilton or A.Wainwright as my last keeper. I’ll probably wait to so how he he does in spring before making that decision.

    • http://www.baseballprof.com/ Bryan Curley

      I think I would still go Wainwright. While Hamilton’s SB are enticing (and valuable as the video shows), they do tend to be more readily available. A legit ace that throws 200+ IP is hard to come by.

      Then again, 100 steals is possible!

  • Connor

    Thank you very much. That was a great video

    • http://www.baseballprof.com/ Bryan Curley

      Thanks for watching, Connor!

  • Wes Stollings

    I by no means have the intelligence (or time) do a proper analysis to strengthen or argue your point, but I have one concern with the method used. You used 300 PA as the baseline to determine how many of the players on your roster had less than 2 SB, however 276 players met that PA requirement last year. 276 players (batters) would not be rostered in an 11 team fantasy league. Based on the same math that you used, by filtering out the bottom 150 PA guys, leaving you with 12-13 rostered “hitters”, where would the significant drop be? It seems to me that by increasing the average players number of steals the value of the steals decreasing would have to occur earlier or later. This analysis is probably accurate for a 20 team league where all these guys would be owned, but I think you could strengthen your argument by using 130 players, assuming the average SB totals would increase.

    • http://www.baseballprof.com/ Bryan Curley

      Thanks for the comment. You are certainly correct that 276 players would not be rostered in 10-16 team leagues, but if a player is looking to add steals to his roster, they will probably expand their pool of players they consider adding to include all 276. By looking at the top 150 or so (whether in terms of PA or fantasy rank) you exclude a lot of valuable players who can provide stolen bases. In a H2H league for example, with one week samples, SB only players that are low in the rankings or who might not tally many PA for the season actually can come up big. When there’s a lot of these guys in the league, players like Hamilton get devalued.

      In the end, it’s the ability to find stolen base production that we want to look at. And don’t forget, a player with 300 PA can be valuable because they might play everyday for half a season. I also used SB/PA instead of total SB to put all players on a level playing field regardless of PA, and after half a season I’d expect the SB/PA rate to be rather stable. I hope that answers your question!