Points vs rotisserie

An intrepid reader, Dave, sent me a great question and comment about the relative merits of points leagues and rotisserie leagues. He felt, and brief investigation agrees, that points leagues are often left out of the fantasy discussion despite the fact that different players and strategies are recommended for each.

A points league rewards stats with points and then the owner with the most points wins. I’ll quote Dave’s league: “Our points league scoring is pretty standard. We did some tweaks to reward OBP, which basically means that we deduct for ABs, but it’s 1 point for a single, 2 for a double, 1 for a RBI, 1 for a run, 2 for an SB, 1 for a BB, etc. For pitchers, it’s 5 for a W, 5 for a QS, 10 for a S, 1 for a K, 1 for each out, -1 for each walk, hit, or ER, and -5 for a loss.”

Which type of league, points or rotisserie, is preferable is largely up to you and your league. A points league provides two big differences: players’ values change, and a team’s overall ranking depends a lot less on the performance of other teams.

To see how player values change, let’s turn to a simple trade-off scenario. Normally, when economists discuss trade-offs, we use a beer versus pizza example and the end result is that you trade some of your beer for some of your friend’s pizza so that you have some of each and not all of one. If you must know, this is due to something called Diminishing Marginal Utility—the more you have of something, the less you want an additional piece. Eventually, you’ve had so much beer that you’d prefer to have a first slice of pizza than an additional beer. A rotisserie league is like a beer and pizza league because if you lead in, say, RBIs, and are behind in stolen bases, you’re likely to want to trade some of the former to get some of the latter.

In a points league, to abstract for a second, there is no diminishing marginal utility. It is more like Miller Lite vs Coors Lite—you like each one equally and you’ll buy which ever one is cheaper. No matter how much Coors you already have, you’ll keep buying Coors as long as it is cheaper. In a points league, you’re not going to trade RBIs for stolen bases just because you don’t have many stolen bases.

What are the practical effects of this difference? In a points league, a trade of like-for-like, such as one outfielder for another, should only happen because the two owners involved in the trade have different expectations about these outfielders’ future performances. Otherwise, there’s no reason to trade an Adam Dunn for a Carl Crawford.

A points league also gives the commissioner more control over the relative value of different stats. In rotisserie, each chosen stat is equally valued. So, if your points league wanted to discourage the common, all-reliever strategy but still keep saves as a stat, you could just down weight the number of points that saves get. Of course, if you’re not careful, you could end-up over-rewarding a particular stat, thereby encouraging teams to fill up on, say, home run hitters.

Also note that in points leagues, your strategy is more independent of other owners’ strategies. The value of a particular stat to you is independent of how much of that stat you think other owners are going to get (however it is not independent of the availability of that stat on the waiver-wire). So if a lot of teams are drafting or auctioning closers, the remaining closers are not more valuable than they would otherwise be. In a rotisserie league, scarcity of closers can drive up the price of saves.

Lastly, points leagues provide more avenues for catching up to the leading teams in your league. Furthermore, tanking teams no longer hurt your chances of doing so. To catch up, all you need to do is get sufficiently more scoring stats than the leading team (and any other team ahead of you). It doesn’t matter which scoring stat you do that in though.

To summarize, points leagues mean less scope for trades, more control over the value of particular stats, less scope for strategy and more opportunities for late-season comebacks.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: Three things I learned in Vegas
Next: Fantasy fallout: Kerry Wood to close for Indians »

Comments

  1. NoPepperGames said...

    I’m also in a points league, but I have to say that I’m not a big fan (but not as much of an anti-fan as I am to Head-2-Head leagues). 

    I have to wonder how ridiculously overpowered closers are in reader Dave’s league are.  One year we had saves at 10 pts (the same as wins), and loading up on closers pretty much sealed my win by July.  We knocked them down to 5 pts the next year.

  2. D Wrek said...

    Great article.
    Im going off of a small sample here, I apologize.  I was in a points league once and felt it very difficult to catch up.  I was pretty much stuck in my position all season.  Like you said trades arent as valuable and the waiver wire is what it is (just more points).  Am I supposed to trade a guy getting points for another guy getting points?  Seems like a wash.  Small sample, but it seems like its all about the draft and dilligent waiver wire work.

    However roto leagues offer more avenues to improve via trade.  You have a huge lead in HR and RBI, trade some of that power for SB’s and SV’s and you could gain a lot in 2 more positions while hopefully only dropping a little in the other 2. 

    You can cover for a bad draft in a roto league via trade, but you really cant in a points league.

    Though like I said, Im working off a small sample of points lagues.

  3. Patrick DiCaprio said...

    I have played in a 12 team mixed head to head points league for about five years, and that format really ratchets up the luck factor. the end of season points champ usually has around 8,000 and the last place team is usually around 7200 or so. it makes things interesting on a year to year basis and even the weaker owners have a good shot every year; for good or ill depending on your point of view.

    You are 100% right about the tradeing issue. In fact in this league we have only had about one or two trades a year!

  4. JC-L said...

    In my experience, it’s been easier to come from behind (or, unfortunately, to be overtaken) in a points league. Think of it this way: You’re in a twelve-team rotisserie league, and the guy you’re chasing has the home run lead and you’re only getting 8 points in that category. As players drop out (which happens even among “serious” leagues), the league splits into two halves. You know that there are 6 players who won’t pass the league leader in home runs, so you can’t get help from them, so you need the other players already ahead of you to help you beat the person most ahead of you.

    In a points league, you’re not reliant on other teams passing other other teams, nor do you need any team to play poorly, as long as yours plays better. You can keep banging out singles to pass people, racking up points there that, in a rotisserie league, might not be valuable in a category you’re doing well in (like BA or OBP).

    You’re right on, though about the problem of trading in order to gain ground.

  5. Pete Schneidler said...

    Has anyone done a points league and attempted to use the actual values of the different events?  For example, a stolen base and double are usually both 2 points, but we know that they are not equally valuable in the real game.  And there are numbers detailing the precise value of just about every event, based on historical averages.  It seems like such a simple idea but I’ve never heard of anyone doing it…

  6. NoPepperGames said...

    Pete, I’d like to hear more as well.  I thought of something walking to my car from work yesterday (when it seems I get the most thinking done…) – how exactly does Reader Dave’s league work?  Are they really valuing a walk more than a single (if they’re subtracting for an AB, and adding a point each for a single and a BB, then the BB is worth more)?  Is that justified, do you think?

  7. Derek Carty said...

    I’ve never played in a points league, but I wanted to jump in and say great article, Jonathan, in addition to some great comments as well.  I think I’ll always be partial to roto, though I like the thoughts about making up ground in points leagues.  One of the biggest drawbacks of roto is people dropping out mid-year and having it affect the standings.  I’ll definitely have to try a points league one year.

    Also, Pete, I had the exact same thought as you while reading this article.  That would be a very cool setup.

  8. cannatar said...

    I’ve been doing a points league since the mid-90s. I’ve tweaked the scoring system a bit over the years and it could be tweaked a little more, but it’s pretty close to “correct” values: 1 for a hit, 1.66 for a TB, 2 for a BB/HBP, 1 for a SB, -1 for an “out” (AB-H+CS).

    Pitching is a little trickier. I use 2.75 for an IP, -4.5 for an ER (works out so that a 5.50 ERA = 0 points), 1 for a K, -1 for a BB, 1 for a Sv.
    There’s an argument that you only need to use IP & ER, but I like adding in K&BB;to both reduce the influence of team defense and ballpark (park factors for BB&K;tend to be lower than or overall runs). I give a point for a save as sort of a cheat to give closers a boost for the added leverage in the 9th inning. The system results in a pretty good balance between hitters and pitchers, and between starters/closers/middle relievers.

    At least from the sabermetric view of baseball, a good points system is much more accurate than rotisserie. It does cut down on trades and the number of strategies, but I prefer it because it seems close to how real baseball works – GMs have one basic strategy: acquire the best players.

  9. Jonathan Halket said...

    These are great comments.  I had actually thought about including a paragraph on some ways to set up point systems, but ultimately didn’t because I realized I could write a lot about it. I’ll write some about it in my next article.

  10. Dave said...

    Jonathan, thanks for writing on this topic. I’m the original Dave who emailed you, and wanted to address some of the comments here. This might be a long post, so please bear with me.

    Of course, my original motivation for writing to Jonathan is to get more recognition for points leagues overall. I love reading fantasy baseball analysis, but find it overwhelmingly skewed towards rotisserie, despite CBS reporting that (I think) around 10-15% of leagues are points-based.

    At any rate, let’s dive in and address some of the topics in the comments:

    - Trades

    As I mentioned in my email to Jonathan, we’ve had to tweak rules to encourage trading, but last year we had 25 trades. Is that above or below roto average? I’m not sure.

    However, in speaking directly to Jonathan’s point about “verisimilitude” in his original article, what drives trades is essentially position scarcity. We have larger rosters and position limits, so you can start 162 games at 12 hitting positions, and then 162 games each at SP and RP, with 27 players on your team, plus 2 injured and 2 minors spots. If you wind up with depth, it is almost always beneficial to trade from that depth. You’re not trading just “points for points”, you’re trading “2B points for SP points”. I think this is a lot closer to real baseball.

    We very rarely see “challenge trades” of 2 players at the same position.

    Also, see my comment on keepers below.

    - Catching up

    While I haven’t spent a ton of time in rotisserie leagues, last year our champion was in last place around early summer and won the whole thing. This year, I was in last place in May and finished second. I’m not sure how common such comebacks are in roto.

    Yes, of course it’s possible that you’ll fall way out of it. Fortunately, we compensate for that with…

    - Keepers

    One critical piece I mentioned to Jonathan is that our league has 4 keepers that are contract-limited to be kept for only 3 years. The contracts are a relatively recent rule change that have encouraged trading and participation, and make the draft quite exciting. This year, at the deadline, we saw trades of Utley, Oswalt, Webb, Rollins, and Lincecum, to name a few. Clearly, though, keepers aren’t just limited to points leagues – maybe Jonathan will write about the contract idea in a future article.

    Furthermore, keepers are a very nice incentive for those that are out of it to keep playing. Like I said, I’m not sure if it’s easier to catch up in roto or points, but keepers allow those that can’t catch up to stay involved and prepare for next year.

    We (finally) have a league in which all 12 owners are very involved, but the “deadbeat factor” that Jonathan alludes to above is critical. If someone tanks, then it doesn’t affect anyone else in a points league.

    - Luck

    Any head-to-head league has a large luck factor, whether it’s roto or points-based. I’d argue that roto-based-weekly-h2h is akin to flipping a coin 26 times during the year. Points-based weekly at least smooths out the “lucky wins” or “lucky stolen base” factor.

    I don’t think anyone wins a non-h2h league via luck if it requires daily transactions over a 6-month season.

    - Point values

    My summary for Jonathan was a bit simplified, so I apologize for any confusion. First of all, we continually tweak the points in order to try and get hitters and pitchers to be valued equally. It used to be that due to the dearth of ace pitchers and our keeper rules, no ace pitcher was ever available. I think cannatar’s summary / points system sounds good as well.

    We tweaked the rules to increase hitter values last year. Here are the real values for important categories:

    H: .75
    1B, 2B, 3B, HR: 1-4
    AB: -.25
    BB: 1.5
    CS: -1
    GIDP: -1
    KO: -.75
    R: 1
    RBI: 1
    SB: 2

    You’ll notice that a single is: H + 1B + AB = .75 + 1 -.25 = 1.5. A walk is also 1.5. This is not a coincidence.

    (And I know that subtracting for a K is a bit controversial…)

    As for relievers, first of all, you are game-limited for relievers, so it’s tough to “load up” on them, especially since you can only start 3 at any time. Very rarely will anyone have 3 startable closers. Furthermore, there are significant risks associated with closers, since our scoring is as follows:

    S: 10
    W: 5
    Relief Win: 5 (to cut down on the bad luck of blowing a relief appearance on a tie game)
    Blown Save: -5
    Loss: -4 (we actually lowered this from 5)
    BB, ER, H: -1
    K: 1
    IP: 3
    Hold: 7

    We added holds a couple years ago to allow a backup plan to the person who gets screwed by tons of injured closers and closer-by-commitee situations.

    - Additional notes:

    Jonathan says: “So if a lot of teams are drafting or auctioning closers, the remaining closers are not more valuable than they would otherwise be.”

    This isn’t true. If a lot of closers are going in the draft, you better get yours or find some sleepers, because otherwise you’re going to have 162 relief appearances wasted with very little recourse for catching up on them.

    As another note, here’s a list of the top 10 hitters in points last year (note that of these, I’d argue that Wright or Pedroia might have been the most valuable, due to position scarcity…but probably Pujols tops the list.)

    Pujols, Albert: 687.2
    Berkman, Lance: 622.2
    Wright, David: 621.8
    Beltran, Carlos: 616.0
    Sizemore, Grady: 610.0
    Reyes, Jose B.: 607.0
    Teixeira, Mark:  604.5
    Ramirez, Manny: 598.2
    Pedroia, Dustin: 598.0
    Utley, Chase: 594.5

    Another interesting tidbit I’ll point out (at the risk of players in my league reading this) is that due to the glut of roto-based advice, it’s almost always the case that players that get steals in our league are overvalued (Taveras, Gomez), as are
    HR hitters that K a lot (Howard).

    Thanks for the feedback, guys. I appreciate the write-up and the thoughtful comments and would love to discuss this more…why don’t you all join a points league next year and see how much better it is! (And slowly turn the tide of fantasy commentary…)

  11. Mark Evans said...

    Pete Schneidler said…
    “Has anyone done a points league and attempted to use the actual values of the different events?” 
    The points league at STATS.COM was originally developed by Bill James with realistic relative weights for all the events. This game went dark for several years when MLB was throwing its weight around to turn fantasy baseball into a cartel, but it’s back in business now.

  12. D Wrek said...

    Good stuff Dave.  The league I was talking about was cumulative over the whole year, rather than a weekly h2h.  As I mentioned, I didnt like cumulative, but could see myself enjoying the weekly H2H.  Definitely agree that roto based h2h leagues involve a lot of luck and I do see how points based h2h could eliminate some of that luck.

  13. John said...

    I’m involved in a 12-team (mixed) keeper league, based on points.  We’ve been tweaking the points system for 6 years, and last year we finally settled on something that lines up very well with runs created and pitching runs created.

    We just couldn’t stand the thought of being one of those leagues where Juan Pierre was being taken in the top 5 rounds, and Chase Utley was being traded for Brian Roberts because the guy who owned Utley “had enough homeruns, and wanted more steals.”

    Imagine actual baseball teams were run that poorly?  What a joke.

    I don’t like the idea of leagues being scored on line-drive % or component ERA, but if you can find a way to balance out the “standard” baseball statistics in such a way that sabermetric ideas are still somewhat represented, you’re league will be in great shape.

    The league winner should be determined by who makes the best decisions with their keepers and in the draft, not by who did the best job of scouring the waiver wire to balance their team’s HRs and SBs.

    Then again, maybe I’d feel differently if I participated in re-draft leagues.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>