Every winter, a couple of people using complicated formulas project how every player in the major leagues and quite a few minor leaguers will perform the following season. It's fun to look at and talk about, and it gives us a thumbnail sketch of what we might be able to expect the team to do if we sub in or out certain players.
When we look back at how players did when the season is over, we're left with the question of the chicken or the egg, though. Did the players overperform or underperform what they should have done? Or did the projection just get it wrong? It's hard to say.
In any case, I thought it would be interesting to look back at what Dan Szymborski's ZiPS projections at Baseball Think Factory said some Tigers should do in 2010, vs how the season really turned out. I tried to grab a cross-section of four starting position players and five pitchers to do this. All but two were with the Tigers last year. The other two were Yankees.
Miguel Cabrera:
The projection system gave a pretty solid endorsement to Cabrera in 2010, but he managed to blow past it in one of the best seasons in history by a Tigers first baseman. Hard to fault it too much I guess.
| Average | On-Base Percentage | Slugging Average | HR | OPS+ | |
| Projected | .305 | .376 | .536 | 34 | 141 |
| Actual | .328 |
.420 |
.622 |
38 |
179 |
Alex Avila:
It appeared as if ZiPS had a pretty low opinion of Avila. Turns out thanks to some bad luck on balls in play, it was actually a bit too optimistic, but pretty close overall.
| Average | On-Base Percentage | Slugging Average | HR | OPS+ | |
| Projected | .239 | .309 | .372 | 8 |
82 |
| Actual | .228 |
.316 |
.340 |
7 |
79 |
Brandon Inge:
With Inge coming off a year where he tried to play through an injury and had a horrible second half, the system saw only the bad and possibly weighted his earlier poor seasons while playing catcher a bit too heavily. So he overperformed expectations slightly. I would guess its next projection will be a bit closer to reality.
| Average | On-Base Percentage | Slugging Average | HR | OPS+ | |
| Projected | .226 | .308 | .390 | 18 |
86 |
| Actual | .247 |
.321 |
.397 |
13 |
94 |
Austin Jackson
With only minor league stats to go off and no clue just how well Jackson would drive the ball, ZiPS completely underestimated what he brings to the team. Notable is that its isolated power projection of .092 wasn't too far off from the .107 reality. I think we'll see it understimate Jackson again next year too.
| Average | On-Base Percentage | Slugging Average | SB |
OPS+ | |
| Projected | .245 | .296 | .337 | 16 |
70 |
| Actual | .293 |
.345 |
.400 |
27 |
102 |
Pitching
Justin Verlander
ZiPS foresaw Verlander taking a slight step back across the board from his phenomenal 2009 season. He actually had a better ERA in 2010. It was remarkably close otherwise.
| ERA | Innings | Strikeouts | Walks | ERA+ | |
| Projected | 3.90 | 219.1 | 217 | 75 | 118 |
| Actual | 3.37 |
224.1 |
219 |
71 |
124 |
Rick Porcello
Again, a rather close projection by the system. It was nice to see Porcello show better control than he was expected too and have more strikeouts. We'll have to see what his comparisons look like next year.
| ERA | Innings | Strikeouts | Walks | ||
| Projected | 4.89 | 151 | 73 | 55 | 95 |
| Actual | 4.92 |
162.2 |
84 |
38 |
85 |
Jeremy Bonderman
I think one problem ZiPS has is trying to make sense of a player coming off an injury-shortened season. We can see it actually did pretty good on projecting the rates for strikeouts and walks, but it was too optimistic on the ERA.
| ERA | Innings | Strikeouts | Walks | ERA+ | |
| Projected | 5.12 | 89.2 | 61 | 35 | 90 |
| Actual | 5.53 |
171 |
112 |
60 |
76 |
Phil Coke
| ERA |
Inn |
Strikeouts | Walks |
ERA+ | |
| Projected | 4.91 |
69.2 |
57 |
29 |
93 |
| Actual | 3.76 |
64.2 |
53 |
26 |
112 |
Ryan Perry
Perry's inabilty to keep from giving batters a free pass in his first full professional season left the system with little to go on in predicting 2010. However it did see an above-average pitcher and got the ERA+ quite close.
| ERA |
Innings |
Strikeouts |
Walks |
||
| Projected | 4.42 | 77.1 | 72 | 46 | 118 |
| Actual | 3.59 |
62.2 |
45 |
23 |
117 |
Conclusions:
Projection systems will never be perfect. It's not quite like trying to hit a bullet with a bullet, but it's something akin to it. Even a good assessment of skill level of a player can be wrong because of the good or bad luck and other unforeseeable event that go on during the season.
Personally, I like to take a look at several system and mesh them together rather than rely on any single one. But ZiPS is a favorite of mine to look to when I'm in a pinch, and easy to find on Fangraphs.com.
When the latest Tigers projections are announced, you can bet we'll take a look at them here.
Source:
Actual stats (OPS+, ERA+) obtained at Baseball Reference.