On a podcast I listened to yesterday there was some interesting discussion about Albert Pujols and the fact that most experts were wrong about him.
On the debut show of the Fantasy Bonus With Adam Ronis show Adam and his guest Paul Greco were discussing the Albert Pujols situation and the fact that most of the experts were wrong about him. I am one of those “experts” if I can be bold enough to use their term, as I recommended against taking him because of the injury risk.
The item that readers must recognize is that Pujols’ success means nothing in terms of impugning the decision making process of this recommendation. As we continually say here, and as we have enshrined as one of our Core Principles:
Perhaps most importantly, decisions can only be evaulated based on information that existed at the time of the decision. Do not let results seduce your evaluation of the actual thought process. Process wins over results in evaluation.
The Pujols situation is a perfect example. Without reciting all of the evidence that existed at the time of that decision, including statements by Pujols’ himself, it is beyond any serious argument that the decision to pass on Pujols by anyone who made that call in their draft was the correct decision.
Let’s assume for argument’s sake that Pujols would be worth $35 as a full season projection. You pass on him and get at least average value out of another first rounder, and let’s say that the average value of the pick is $31. So by omitting Pujols you incur a loss of $4 based on his full season projected value versus the player you selected.
Let’s also assume that if the statements made in the spring that Pujols would have to have surgery by July (and let’s use the end of July as our yardstick for simplicity’s sake) were true, then we can assume that if he missed those two months and you selected Pujols you would have lost approximately $11 when he went out.
Given these assumptions, it follows that you are risking a loss of $11 in pursuit of that $4 profit. So what is the break even point? You have to earn that $4 almost 75% of the time to break even. To put it another way, was there an approximately 25% chance that Pujols would go out in July? Given the information that existed at the time the answer is clearly yes.
Like all decisions, we must evaluate the process rather that the result. When we are dealing with contingent events like this we are weighing probabilities and making the play with the highest expected value. Whether the actual result in this one “trial” of the proposition is a winner or loser is utterly irrelevant. Like in poker, sometimes the best hand doesn’t win; that doesn’t mean you played your hand incorrectly.
Another interesting point raised by Paul Greco was the allegation that anyone who says they “read between the lines” of what was said and divined that Pujols wasn’t really hurt too badly is lying. I think there is a lot of truth to this allegation; in fact it is almost certainly true.
It is very easy to come up with an after the fact justification for a result; this is exactly one of the issues raised by the Texas Sharpshooter fallacy. If someone shows me where a particular wrietr or analyst said at the time that they read between the lines and predicted that Pujols would play the full season then I would love to see it and would give that person full credit for a great decision making process. Until I see that I side with Greco on this and would readily assert that they are lying.
Paul’s point is that many so-called experts did not actually consider the issue properly. In fact, in the absence of an opinion that Pujols would play a full season one could legitimately argue that Pujols’ mere presence on an expert’s team means one of two things: either they did not evaluate the situation correctly or they took him so that they would have an excuse if the team failed.
Paul believes that it the second is true based on his history and experience in the industry. Given my relatively limited history among fantasy baseball writers I think it is mostly a matter of the first. Very few analysts weigh risk and reward correctly and this is immediately obvious from most of what is written out there. You can count on one hand the number of bloggers that write about fantasy baseball that will discuss the issue in stark and discrete enough terms that you know they are weighing the considerations properly.
So when you hear someone touting the fact that they picked Pujols keep these considerations in mind.
6 responses so far ↓
1 Brian Joura // Sep 4, 2008 at 1:29 pm
As an owner who had the #2 pick in one league and never considered Pujols, obviously I agreed that the risk was not worth the reward.
But at the same time I don’t think it’s fair to say that you evaluate decisions on the process rather than the results.
Results count and to ignore them completely is off-beat. This sounds a little like the scouts vs stats thing. Why not use a mixture of both?
I wouldn’t blame anyone drafting in the very bottom of the first round from picking Pujols this year. While rumors of surgery were out there, this was also a guy who never played less than 143 games in his previous seven seasons. Rumors about Pujols and surgery don’t carry the same weight as rumors about Nick Johnson and surgery.
If someone wants to crow about picking Albert Pujols I’m quite willing to let them. I’d much rather hear that than some guy trying to tell me he knew Cliff Lee was going to win 20 games this season.
2 Mike Podhorzer // Sep 5, 2008 at 7:09 am
I wish more people understood this concept. It would certainly cause me to get into fewer arguments! I will always believe that process is more important than results.
Brian, I don’t think Pat’s saying results aren’t important, but the point being that if your process was faulty but ended up with positive results in spite of this, then you shouldn’t truly be deemed a success, but rather lucky.
As a crazy example, if I projected a breakout for Carlos Quentin and my process was that I knew he bought a new house before the season and he was therefore very excited and happy in life which would motivate him to perform well this season, then I shouldn’t get any credit for predicting the breakout because my reasoning was awful, so I merely got lucky.
Another example would be starting a pitcher who ends up pitching very well in a daily transactions league because you had a “gut feeling”, when all the stats in the world suggested it was a bad idea. The process here was poor, so the results shouldn’t matter.
3 Brian Joura // Sep 5, 2008 at 7:41 am
On the specific example of Pujols, there was risk involved and it was up to each individual owner to assign how much risk it actually was.
My opinion is it was not a case where you could put an exact dollar amount or precise figure of any kind on. Others might disagree.
My feelings were that Pujols was too risky to take in the first half of a draft. Personally, I wouldn’t have taken him before #10.
But I think this is a case where you have to look at results to get the full picture of how to justify your decision.
Picking Pujols in the first round this season was a risk. But you have to take risks to have a good team (or just get lucky, I suppose).
In the Pujols case I’m quite willing to congratulate the owner who assessed the situation, determined that the risk was not that great and rolled the dice with him.
I think there was enough information available to determine it was a tolerable risk to take. Pujols is a durable player, his manager loves him and will put him in the lineup if possible and when was the last time you heard a position player in his prime say before the season that he likely wouldn’t play the entire year and have that actually come true?
It’s not a good idea to take nothing but safe picks throughout a draft. Pujols was not a safe pick and it turns out he rewarded owners who took him.
No matter what style you play, you win some and you lose some. And this is certainly a case where an individual could have made an informed decision, taken a risk and have it pay off.
4 Patrick DiCaprio // Sep 5, 2008 at 7:44 am
Sounds like a face-off topic…
Mike you have it right; I am not saying results are unimportant in general.
5 John // Sep 5, 2008 at 8:27 am
I think most leagues implicitly factored in Pujols’ injury risk. In one auction league I play in, he was the 10th highest NL player (at $32 - tied w/ Peavy and Dunn at that price). Without the news of a potential injury, he would have never went that low. I would think it’s very hard (if not impossible) to quantify the “risk/reward” ratio for this situation.
6 Patrick DiCaprio // Sep 5, 2008 at 1:46 pm
John the discussion was really about draft leagues. in auction leagues you are right; there is no risk/reward because price speaks. That is data point 186 why i hate draft leagues and love auction leagues.
Leave a Comment