Steamer and I: Yasiel Puig
It’s back to another Steamer and I, as I compare my Pod Projections to the Steamer forecasts and identify players we disagree on most. Last time, I switched gears, discussing Mike Trout, who surprisingly I was far more bearish on than Steamer.
I am continuing the bearish theme with our next man, the ManBearPuig himself, Yasiel Puig. Puig made a splash in his 2013 debut, posting a .398 wOBA with strong power, some speed, and excellent defense in right field. But since, his offensive production has been in decline and last year he battled injuries en route to just a .328 wOBA. I think we all assume a rebound is in order, but the question is how much?
For our projection comparison, all 2015 and Steamer counting stats have been extrapolated to the same number of plate appearances I forecasted.
| System | PA | 2B | 3B | HR | BB% | K% | AVG | OBP | SLG | ISO | wOBA | BABIP |
| 2015 | 567 | 22 | 5 | 20 | 8.4% | 21.2% | 0.255 | 0.322 | 0.436 | 0.181 | 0.328 | 0.296 |
| Pod | 567 | 28 | 5 | 20 | 8.8% | 20.8% | 0.274 | 0.344 | 0.468 | 0.194 | 0.350 | 0.320 |
| Steamer | 567 | 29 | 5 | 22 | 9.3% | 19.3% | 0.287 | 0.361 | 0.493 | 0.206 | 0.367 | 0.326 |
As usual, let’s move from left to right. Our doubles and triple projections are nearly identical, so clearly this is not the cause of our disagreement. We find a marginal difference in projected home runs and some of that may be how much we believe in the fly ball rate spike Puig experienced in 2015. His FB% has risen from 30.7% to 33.4% to 38.7% over his three seasons, so we have to assume to regression back to his career rate.
Puig’s xHR/FB rates have been excellent, though he has significantly underperformed in the last two seasons. Dodgers Stadium is neutral for right-handed home runs, so that shouldn’t be the explanation. It’s either just been rotten luck or something Puig is doing, or not doing, that the equation has failed to account for. I don’t know which it is, but I’m projecting a slight bump in HR/FB rate to 14%, which he has only cleared in one of his three seasons. I wouldn’t argue with a more aggressive projection, which might be driving Steamer’s home run forecast.
Moving right along, we find that Steamer is projecting a slighter higher walk rate and a lower strikeout rate. These aren’t major differences, but small differences here and there add up. That 9.3% walk rate projection from Steamer is exactly Puig’s career rate. But, his career suggests that maybe his 2014 mark of 10.5% is the outlier and he’s more of an 8% to 9% guy.
But let’s look deeper than just his historical walk rates. His Swing% is crazy high! In 2015, he ranked 27th among hitters with at least 300 plate appearances in Swing%. In 2013, he was 15th highest among those with at least 400 plate appearances. He swung less often in 2014 when he posted the 10%+ walk rate, so we’re left wondering whether he could display that type of discipline once again.
The strikeout rate projection represents a bigger gap than the walk rate. My projection is Puig’s career average, while Steamer is projecting a career best mark. No other projection system is below 20.1%. The thing is, Puig swings and misses a lot. If he didn’t swing so often, his strikeout rate would be significantly worse. So it’s hard to project both marks improving so drastically as Steamer is. If Puig swings less, his walk rate rises, but his strikeout rate will likely do so as well.
We then note a small difference in ISO, fueled by those very marginal differences in doubles and home runs. By itself, it’s not a big deal, but combined with the better walk and strikeout rates from Steamer, it adds up to a meaningful difference.
Last is the BABIP. Puig’s BABIP sat at .383 during his rookie campaign, then it dropped to a still inflated .356, before plummeting to just .296 last year. That .296 mark was actually below the league average. Was he just fortunate in 2013 and 2014, or was his 2015 mark actually the fluke? Well, let’s ask xBABIP!
Hmmmm, so perhaps those seriously inflated BABIP marks during his first two seasons were the result of some great fortune. He doesn’t hit many line drives and fails to go the opposite way on a frequent basis. He does have power and speed though and doesn’t have a pop-up problem, so he appears to have a true BABIP skill that should sit above the league average. But certainly nowhere near his .350 career average. Steamer and the rest of the projection systems are in general agreement, though I’m the low man on the BABIP totem pole.
Once again, in isolation, our BABIP differences aren’t a big deal. But thanks to a whole lot of small differences, it adds up to one of the biggest discrepancies between the Pod wOBA and Steamer wOBA projections.
Given some of the skills he has shown in the past and his exciting tools, it would be no surprise if Puig made me look silly. Heck, it wouldn’t be shocking to see him blow past all projections. We have seen the upside and the downside, so which one shows up this year?
—
So which projection do you agree with more, Steamer or Pod?
Mike Podhorzer is the 2015 Fantasy Sports Writers Association Baseball Writer of the Year and three-time Tout Wars champion. He is the author of the eBook Projecting X 2.0: How to Forecast Baseball Player Performance, which teaches you how to project players yourself. Follow Mike on X@MikePodhorzer and contact him via email.
I’m betting on the come, primarily b/c his draft price is so reasonable (NFBC ADP of 65). If we had to pay top-3-round prices, no thanks. But a 5/6th round pick seems like mostly upside.
Agreed, I think he’s become reasonably priced, which made me surprised that I was actually pessimistic compared to the projections.
If your projections are down on him, but you think he’s fairly priced with upside….do you personally end up dratfting Puig? I guess the question is, how heavily do you rely on your projections when actually drafting?
I didn’t even know I was down on him until I did this exercise! I thought I was actually optimistic about a rebound. I fully trust my projections, since I do all the numbers manually. I’d change them if I didn’t trust them. But due to differences in valuation with the huge number of outfielders we need to draft, there will always be someone I feel is severely undervalued that I draft instead. So I’m unlikely to be drafting guys I feel are fairly valued.
So the challenge is, you know your projections are solid as a whole, but on individual players, are really hit or miss. It’s the nature of all projections. Even if yours are really good (and I hear they are), each player will almost certainly be “off” your projections.
Of course you have to use projections & the resulting valuations as a base for valuing players. But in my mind, it’s a mistake to draft solely based on projections without taking things like plausible upside at draft position (Puig) into account.
Well projections are supposed to already factor in downside and upside and essentially represent the 50th percentile, like PECOTA. So if a player has more upside than downside to my projection, then I’m doing it wrong! It’s why I, and projections, generally don’t project breakouts.