How well do cold streaks predict the future performance of hitters?
Last week in this space, I looked at how well hot streaks predict the future performance for hitters. That is, given that a hitter is currently on a hot streak, how likely is it that he’ll continue that hot streak in the next game? Well, last week I found the answer to this question to be “not very likely.” Last week’s results showed that we’d be better off using a basic, preseason Marcels projection than trusting the hot streak to continue. This week, I thought we’d look at cold streaks and see if we find a similar result or something different.
I’ll be using the same methodology as last week, so I’ve just copied and pasted it here for convenience.
To study cold streaks, I’ll be comparing the stats from the hitter’s 7-day streak to the following day’s stats (day eight). I’d then compare the hitter’s preseason Marcels projection to the hitter’s day eight stats. This will tell us which is more predictive – a full projection for a player or the streak. When comparing these sets of numbers, I’ll be using average error to study the differences.
I looked at all possible 7-day streaks for all hitters back to 1993 that occurred in either March, April or May (I did this so that the preseason projections would be as accurate as possible – if we’re looking at streaks in August, it’s likely the player’s projection would have changed considerably by that point in the season). From there, I included all hitters who had played in at least five of the past seven days and who had at least three plate appearances on day eight. No context adjustments were made to the data.
“Cold” Hitters – 7 days
Let’s switch things up a bit now and look at cold streaks. This time, everything stays the same except that to qualify as a “cold” streak, a hitter needs to go seven days averaging fewer than 0 FanDuel points per day.
Like we saw with hot streaks, preseason projections reign supreme. It’s interesting to note, however, that there’s just a 0.29 FanDuel point difference between using the cold streak and Marcel – much lower than for even a 30-day hot streak.
“Cold” Hitters – 14 days
Wow. Well here’s an interesting result. It appears that using the cold streak is, in fact, slightly better than using a preseason Marcels projection to select your FanDuel hitters after the hitter has been cold for a couple weeks. We’d still prefer Marcel for BA, OBP, and wOBA, but for FanDuel players, the cold streak does appear to be important!
Granted, our sample size isn’t huge (244 hitters qualify with a total of 8073 PAs accumulated during the streaks), but this is definitely a noteworthy finding.
“Cold” Hitters – 30 days
We don’t get a single hitter who qualifies here as, presumably, any hitter who has been ice cold this long is now riding the pine.
Breaking Down the Results
While we found hot streaks to be largely irrelevant last week, for cold streaks, you’re going to be better off using the streak than a preseason Marcel projection after two weeks (and probably a couple days sooner). This is a highly significant finding, though not a be-all-end-all result.
There are a few reasons why we might have found this. The first is that Marcel is the most basic of projection systems. Using a more complex system like Baseball Prospectus’s PECOTA or THT’s Oliver would certainly narrow the gap. Also, remember that I used a preseason projection. Preseason projections won’t change much by the end of May on the whole, but they will change a little, especially when a hitter performs so poorly as to be put into the “cold streak” bucket.
The final reason that cold streaks are more significant than hot streaks is that sometimes, it really isn’t random variation. Sometimes, the hitter really is having problems mechanically or is having personal problems. These will affect him negatively, but there’s no such thing as a “reverse mechanical problem” that would affect the hitter positively. Hot streaks are going to be pure random variation, while there could potentially be a legitimate reason behind a cold streak.
Derek Carty’s work can also be found at Baseball Prospectus, CardRunners Fantasy Baseball, and DerekCarty.com. He has previously had his work published by The Hardball Times, NBC’s Rotoworld, Sports Illustrated, FOX Sports, and USA Today. He is the youngest champion in the history of LABR, the longest-running experts league in existence, and is a graduate of the MLB Scouting Bureau’s Scout Development Program (aka Scout School), one of just two active fantasy analysts to have done so. He welcomes questions via e-mail, Facebook, or Twitter.
- How important are hot streaks for hitters? (fanduel.com)