First off, I would like to say that the best part about writing at Baseball Prospectus is the community and audience that I get to write for. The enthusiasm and intelligence of the readership (at least what I can tell from the commenters) is outstanding. Smart and insightful comments are often the inspiration for my articles and often allow me to better understand the concepts and ideas I am/was writing about. Now they are not all pearls of wisdom, but there ain’t no sweet without the sour.
Anyhow, the three comments from my article two weeks ago, In-Season Advocacy Effect and Managing Multiple Leagues, all brought up interesting points and I figured responding to these questions/points would make for a helpful article (I also apologize for the delay). I figured that if one person is thinking something or asking something, then there are probably others with the same thoughts and questions. At the very least, responding to these comments should help us improve our understanding of these concepts and topics.
Lastly, before we get to the responding to the pseudo-mailbag, my hope is to do something like this every 2-3 weeks. So if you all have any strategy, process, or behavioral questions (in other words, not specific roster questions such as “should I drop Player X to add Player Y?” or “should I accept this trade offer?” (Craig Goldstein will happily answer those on the Bat Signal)), please send them to email@example.com. Thank you in advance.
Alright, let’s get to the questions and comments and the responses. Please note that when I respond to a comment, I acknowledge that the commenter might already understand what I am saying in my response (hence their comment in the first place), but my response is thus directed at the larger audience.
“Love these articles, Jeff. Always an interesting read.
One of the things I struggle with is when I have a pitcher going in one league and hitters opposing him, either in that same league or other leagues. I often feel it's a lose-lose (either the pitcher does well, so the hitters don't, or the hitters do well and the pitcher sinks). That or praying that pitcher throws well and only gives ups a run or two to the hitter(s) I have.
Obviously if you have no better options you start as you would normally but I'm curious of your thoughts on that. I find that it often bleeds into stupidity, where I "need" a certain player to do well in one league so I will bench the opposing player(s) as if to convince myself that will happen.”
Thanks for the kind words, SadMagistrate. Regarding the lose-lose feeling of having your hitters go against your pitcher, I can say this is something we all feel. This can be such a strong impulse because it is so easy to imagine one side doing poorly at the expense of the other. Conversely, it is less easy to imagine one side being mediocre and the other being good or both sides being good (as what happened with Cole Hamels and Andrew McCutcheon on Wednesday for example). There is also some ill effects of the endowment effect (our tendency to overrate that which we own/possess/”is ours”) showing up here in that we are probably overrating the effectiveness of our hitters going against our pitcher or vice versa.
Really, though, this is an issue about decision framing. The error in framing these roster decisions as they are framed above is that it assumes that every player has to click/produce/be productive every time in order for our teams to succeed. The way we should be framing our roster decision is to ask whether we are making the decisions with the highest expected return. When we do this we realize that the decision of which pitcher to start is independent from the decision of which hitters to start even if they are facing each other. We also realize that trying to have every decision workout actually hinders our ability to always make the decision with the highest expected return.
So to finally answer the question asked, please start your players with the highest expected return, which might very well mean starting your hitters against your pitcher. Just take comfort in the fact that you are giving your team the best odds of winning.
“The fact that you have the same player in many leagues might be the result of the advocacy effect, but it might simply be the result of valuing a player more highly than average in the first place.
Multiple auction drafts would not be very good at testing the advocacy effect, because being the winner at an auction means "I value the player *at least* this much, but I may value him even more." A subsequent higher auction where you pay a higher price for a player may be advocacy effect, but it may simply show that you paid less than you valued the player at in the previous draft. Snake drafts are probably better at this.
Auction drafts might be useful is showing a reverse advocacy effect. If you miss out on a player, you'll think of all the bad things about that player and maybe fail to purchase a player at a price lower than your highest (unsuccessful) bid in the previous league.
Finally, the attention 'bias' may be perfectly rational. Attention is a scarce resource, and making a 'sub-optimal' choice from the point of view of perfect rationality may cost attention that is best deployed elsewhere. Time spent researching that second catcher could instead be spent researching batter matchups, etc. The real lesson might be to find ways to decrease attention costs in making choices–using automated ranking systems or other heuristics, or by playing in leagues with similar rules.”
Thanks for the comment, Kinanik. There are a bunch of great insights in this comment and I am going to try to address them one at a time.
Regarding the having the same player in multiple leagues being the result of being the highest on a player as opposed to the advocacy, I agree that this can definitely be the case. Additionally, having a player on multiple teams is almost always going to happen whenever we play in multiple leagues. The issue is whether we are highest on a player because of the advocacy effect or just because we someone has to be the highest on each player. When we start to have the same player on our team year over year, or across multiple leagues with different leaguemates and/or formats is when we are likely to be getting into being-affected-by-the-advocacy-effect territory. The chances of being the highest on a player across multiple leagues, year after year (especially when there are alternatives—that being other players and other strategies) are not very high, yet we see it all the time (thus why I though the discussion of the advocacy effect would be beneficial).
Regarding multiple auctions not being a very good test of the advocacy effect, I disagree. I agree that getting a player in one league only indicates that you value that player at least that much. That said, the fact that you keep ending up with a player across multiple auctions, when there are (usually) many alternative from which to choose, means that you are not only the high person on that player, you are also not the high-person on any other players at the position. Now maybe you designed your strategy around a specific player, thus making this a different discussion, but if not, we should at least be examining or trying to examine if we are being affected by the advocacy effect.
Regarding the reverse advocacy effect, this is just an awesome insight. I have never thought about this other side of the advocacy effect, but this seems to make a lot of sense. Just as we talk up our players (usually to ourselves) to justify our decision, we talk down the player/choices we did not take; I know I have done this. I also know that we use multiple auctions and drafts to grab players we missed out on, which would combat this revers advocacy effect, but that does not mean that they do not both exist.
Regarding the attention bias being perfectly rational, I have some issues with it. First, I am not really that concerned with it being rational, I am really only concerned with whether it gets in the way with us making the best decisions possible. That said, I do think the advocacy effect gets in the way of optimal decision making. Not having to spend time on a decision simply because we ended up with a player (this being a game, our decisions are always at the mercy of our opponent), makes it likely that we are not optimally deciding how to utilize the scarce resource that his our attention. I do think you are onto something in trying to find a way to avoid this decision making trap, and that is improving our process. I also think you are onto something in looking at our attention as a scarce resource when designing our process. Maybe spending a little bit of our attention on deciding how we should be divvying up our attention across our decisions will be more beneficial than it would appear at first blush.
“I've thought about this often as I'm in several leagues. I think that the price of the player is pertinent. I liked DeSclafani and got him for a dollar in several leagues. So far, so good. But even if I was dead wrong, the cost was minimal. However, for high priced players, the risk is blowing up the whole season across multiple leagues. For example, if you bought Corey Kluber as your ace in every league, it has been an extremely painful April.
On the flip side, if you are able to accurately predict several key breakouts but had them spread out equally among all of the teams (because you didn't want to overly weight them), it would be painful if you finished just out of the money because individually they didn't make an impact in any one league.”
Thanks, dangor. What is described above is classic, pseudocertainty effect; that being that we make risk-averse choices when the expected outcome is positive (high priced players) and risk-seeking choices to avoid perceived negative outcomes (end game players). If Corey Kluber was your best bet to return value in multiple leagues, then all you should be doing is trusting your valuations (now if your valuations are flawed, that is a different issue) and making the best investments possible. Put differently, the risk inherent in any player should already be a part of your valuation; thus, every player should have a value and thus a price you would be happy to pay for them. Now if you do not trust your valuations, then they are not really your valuations; thus, you should probably adjust them as to not invite further bias into your process (in that you would not be working off of valuations and thus probably letting a lot of biases sneak into your process).
Regarding spreading out your choices among your teams, well that is really not what is being discussed with the advocacy effect. If a certain player is going below value in multiple leagues, then by all means keep taking the value. The issue becomes when we have to pay more and more for the players we thin will breakout; often doing so to the point where there is no more upside left or, more importantly, doing so to the point where there are more profitable alternatives.
That is all for this installment of the Strategy and Process Mailbag. I hope you enjoyed it. I welcome all feedback as to the format, how I could make this better going forward, or anything else you can think of. Thank you in advance.
Thank you for reading
This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.Subscribe now