Tuesday, April 07, 2009

How Much Should I Expect From My Freezes?

Anonymous wonders how much he should expect from the guys he keeps:
in deciding whether to protect players, what sort of profit margin justifies a keep? Is it proper strategy to keep any player whose keeper price is below their expected value? This may be a no-brainer, but I want to confirm.
I generally try to keep anyone who is undervalued with inflation. There are two notable exceptions where I might not keep a player who is undervalued:
  1. Players whose salaries are between $1-3 and are projected to earn the same as their salary.
  2. Players who are only "undervalued" because of inflation.
With the first group of players, it is easy enough to intuit why you might not keep this type of player. A bottom of the barrel starting pitcher that you have for $3 who you have projected for $3 might go for less in the auction. If he doesn't, you're more likely to cross him off of your list during the auction than you are to sob into your beer afterward. There is enough variability at the bottom of the pile that you should be able to fill in at the end with a $1 player who might be at $2 or $3 on your sheet. This is especially true for pitchers.

The second group of players is a tougher one to gauge because owners have a much tougher time with this concept.

For this example, lets say that you have both Brian Roberts and Magglio Ordonez on your roster. Roberts is at $33 and you have him projected to earn $29. Ordonez is at $29 and you have him projected to earn $27. Inflation in your league is at 20%.

With 20% inflation, Roberts' inflation par price jumps up to $35 and Ordonez's to $32. Now they both look like keeps.

The only problem with this is that if you keep both you're still taking an actual loss. Inflation or no, your guess is that you're going to spend $62 on players who will earn $56. Inflation or not, your freezes don't have enough strength for you to keep both. It's probably going to be better to decide on one or the other.

This is where the strength and depth of your auction will likely come into play. Ordonez seems like the better freeze at a glance. But after Robinson Cano and Mark DeRosa, the next best 2B out there is Nick Punto. Meanwhile, there are 20 OF out there that you have projected to earn $20+. Back Ordonez goes.

Generally, though, you do want to keep value going into your auction, even if it is "only" inflation value. Curtis Granderson was thrown back in my league at $31 and jumped all the way to $35. The owner who bought him at $35 should have offered something to his old owner, even if it was just a minor league pick, so that he could have spent his $4 elsewhere.

1 comment:

Brett said...

I'll try to write more later, but my first "smell test" is simple:

When trying to decide whether or not to keep a player, I pretend that I didn't, and that he is the first player called nominated in the auction.

Then I ask myself if I would bid up to his current salary.

It's really just asking the same question a different way, but for some reason I think it puts you in a different mindset. As silly as it sounds, as soon as he's off your team, you have less of an emotional attachment to him and can think a little more clearly.

Of course, inflation is tricky - let's say I go through the list, and predict that every player with a salary below his (uninflated) value will be kept. I end up with, say, 25% inflation.

Then say I have a guy at $34 who is worth $32. Inflated, his value is $40, so he should be kept.

But now since I'm keeping him, the inflation rate is lower.

Now say everyone is doing what I'm doing. The inflation rate continues to drop and drop. Eventually some of the guys who are kept because they were below their inflated values are no longer below their inflated values.

So what's the lesson?

1) know your league - the better you can predict who they'll keep, the better you can predict inflation, and thus the better decisions you can make.

2) know that because of what I talked about above, you should assume that inflation can swing a few percentage points one way or the other (most likely down). That's also because your predicted inflation will be based on YOUR ideal opinions. Say you would extend a player by 1 year, but his owner ends up extending him 2 years - his salary is then higher than you predicted and inflation goes down some more.

3) since inflation will likely be a bit lower than you predict, only keep guys who are close to their inflated values if they really "fit" on your team. Perhaps knowing for sure that you have Jose Reyes at inflation par lets you not worry about speed and allows you to better strategize for the auction.