Friday, March 07, 2008

Preparing for My Freeze Date

anonymous asked:
How do you go about your auction strategies for re-auctions or keeper leagues?
Before establishing an auction strategy, I like to prepare for the trading and roster strategy that takes place before the freeze date one week before the auction. I don't have a specific template for this, but below are the general steps I follow.

1) Figure out the bid limits for every player.
The first step is to figure out, down to the last dollar, what I think each player is going to be worth. That includes everyone from Alex Rodriguez down to the last $1 catcher. I want the bids to add up to $3,120 ($260 x the 12 teams in my league) so that I'm not overvaluing or undervaluing certain players by mistake.

If you're not as confident in your own ability to do this, I recommend using someone else's bids (Alex Patton is the guy I recommend, but there are plenty of good to excellent sources out there) as a starting point and then tweaking them based on your preferences and your league's proclivities. For example, most expert bids don't take freezes and the dump culture of my league into account, so I tend push certain rookies' bid limits up $1-3 to adjust for my league.

2) Predict the freezes.
This sounds hard, but if you've been in a league for long enough, you can start to figure out other owners' tendencies and what you think they'll do on the freeze date. Keep in mind that some teams will freeze players at par or over value. In some cases, this will be done to fit a strategy or secure stats at a scarce position, but more often than not it's due to questionable judgment.

After you've done this, go in and figure out what the inflation factor in your league will be. To do this, take the remaining money left to spend and divide it by the remaining bid values in your league. If $1,120 of salary is frozen in a league on players worth $1,320, for example, your league will have $2,000 to spend on $1,800 worth of talent. This will leave you with an 11% inflation rate.

If you're not comfortable with this exercise, use your league's inflation rate from last year. It's an imperfect solution, but it will at least give you a general framework.

I recommend doing this exercise separately for the pitchers and the hitters. I'll explain why later.

3) Calculate player value with inflation
Not only will inflation impact how you will do in your auction, it also will impact the value of your players in a trade. In a league with no inflation, a player with a $40 salary who is worth $40 and a player with a $10 salary who is worth $10 basically have the same value, and you probably wouldn't trade either player for the sake of making a trade. However, in a league with inflation, you want the $40 player. Your 30 auction dollars are going to be fighting an uphill battle, and you'll only get back $23-25 worth of value with a 20-30% inflation rate in play.

To get a more accurate picture of off-season player value, you'll want to calculate hitter and pitcher inflation separately. If you're using a 67.5%/32.5% split like I am, that means that your hitter bid limits should add up to $2,100 and your pitcher limits to $1,020 in a 12-team, $260 league.

You will probably find that your pitching inflation is higher than your hitting inflation. The reason for this is simple: because pitchers are less predictable than hitters, it's more likely that a Fausto Carmona will slip through for $2, like he did last year in my league, and become a $20+ pitcher the following year than a $2 hitter will. More importantly, the smaller pitching pie means that each highly undervalued pitcher will have a greater impact on the pitching prices in your auction.

Why do I believe this is important?

In the old days, I used to figure out inflation using a linear model. If I had Dustin McGowan at $10 and my bid on him was $17, he would be $1 less in value than Alex Gordon, who is also at $10 but has an $18 bid limit. I doubt I'd make this trade anyway to get $1 of value back, but I might have been more likely to be swayed by a cheap throw-in. Then after my auction, I'd wonder why my pitching sucked.

Using a non-linear inflation model means that McGowan is worth about $2 more than Gordon in my league, and I won't make that trade. The math tells me that if McGowan and Gordon were both available in the auction, McGowan would go for more since it's harder to find a pitcher like McGowan than a hitter like Gordon. You can probably intuit this without completing this calculation, but I feel better actually having the math in front of me.

4) Try to make some trades.
Since I almost always field a competitive team in my money leagues, I seldom wind up dumping and coming into the off-season with a killer freeze list. The result is that I need to make some trades to try and improve my team.

The goal at this point is simply to add value. There are some commonsense rules that I follow. If I already have four undervalued outfielders, and someone makes me a superior value offer of an outfielder for a shortstop, I won't make the trade if I think the shortstops in my auction aren't going to be there and there are a lot of outfielders to be had. Generally, though, value is the most important factor in making a deal, and these situations are exceptions rather than rules.

I'll talk about my strategy for preparing for my auction after I've frozen my team in my next post.

6 comments:

Anonymous said...

Hey Mike,

After all these years, still not sure I agree about hitting vs. pitching inflation.

In a keeper league, the amount of money that teams spend on pitching is not fixed. In other words, the 65/35, 70/30, or whatever split applies to the values, but not the total money spent.

Consider this league:

25 players frozen and they're all pitchers. They are worth a total of $450 (in a redraft league) but have salary of $30.

Assuming a 70/30 split and 12 teams there is $486 of value left for the remaining 83 pitchers, and $906 of salary. The pitching inflation rate is thus 86%.

Does this mean that the owners are going to spend at a (aggreagate, regardless of inflation is applied linearly or not) 86% inflation rate just so the "money split" equals 70/30? Of course not. They are going to shift that money to hitting.

Or is that not what you meant by separate hitting and pitching inflation?

-Brett

Dr. Hibbert said...

Hey Mike,

Can we take a brief look at position scarcity in 4x4 leagues (preferrably AL-only)? Thanks.

Unknown said...

Thanks a lot for your input on my CBS message board.

Unknown said...

...one of these days I'm gonna have to try one of these auction format leagues.

Anonymous said...

I guess I should be "anonymous one" from now on. Thanks for answering my question.

Another thing about auctions, though: I would presume that inflation (or deflation) can change from pick to pick as the room loses money -- then as teams slow buying, etc. How do you account for that? In essence, isn't it possible to go into an auction expecting 20 percent inflation but see much higher (or lower) percentages as the auction goes on? Are you figuring inflation just as a barometer for selecting top-tier players, since they are most likely to be called out first?

Also, I assume that it would be a stupid idea to compute inflation after each pick, since you don't want to overbid each round and miss endgame players who fall below projected prices.

Thanks, as usual.

Toz said...

Anonymous (at the bottom; I'll get to Brett in a minute) -

You hit the nail on the head, my friend. Inflation does change from bid to bid in the auction...as each player goes off the board, that player either went over your bid limit or under your bid limit, and, therefore, impacts your inflation calculation.

Following Mike's lead, I run a pretty simple +/- notation on my auction sheet; I don't do as much math as Mike and I'm more of a gut feel kind of guy. Why keep track? Because you want to find the soft spots in the auction where you will get players under your bid limit. Mike set this out in his very nice article on Inflation Par, at http://rotothinktank.blogspot.com/2008/03/inflation-par.html. Plus/Minus and inflation is a topic that could take us a month to walk through, and still make for interesting conversation.

Brett - your overall point is correct, at least anecdotally. It is very difficult to believe that any owner would spend, let's say, $90 on Santana. On the other hand, if you do not calculate splits based upon your normal league spending, then apply separate inflation, you run the risk of your bid limits falling well short of where they need to be for the top tier players, without bargains to fall back on in the middle or late rounds. You will wind up not spending enough money and not buying enough stats.