ISO Risers and Droppers
Every season some players have a power breakout (Granderson) or meltdown (Dunn). I decided to look at the players with the highest and lowest ISO increases this season and the reasons behind the changes for a few of the players.
Below is a list of the biggest gainers and losers in ISO from 2010 to 2011 (min 400 PA each season):
| Name | 2010 ISO | 2011 ISO | Difference |
| Cabrera Asdrubal | 0.071 | 0.189 | 0.118 |
| Kinsler Ian | 0.125 | 0.212 | 0.086 |
| Granderson Curtis | 0.221 | 0.304 | 0.083 |
| Zobrist Ben | 0.115 | 0.196 | 0.081 |
| Molina Yadier | 0.080 | 0.156 | 0.076 |
| Upton Justin | 0.170 | 0.244 | 0.075 |
| Cabrera Melky | 0.098 | 0.168 | 0.069 |
| Braun Ryan | 0.197 | 0.259 | 0.062 |
| Gordon Alex | 0.140 | 0.199 | 0.059 |
| Victorino Shane | 0.170 | 0.226 | 0.056 |
| Wells Vernon | 0.242 | 0.181 | -0.061 |
| Cabrera Miguel | 0.294 | 0.232 | -0.062 |
| Martinez Victor | 0.191 | 0.128 | -0.062 |
| Abreu Bobby | 0.180 | 0.109 | -0.071 |
| Rios Alex | 0.173 | 0.101 | -0.072 |
| Ethier Andre | 0.201 | 0.129 | -0.073 |
| Werth Jayson | 0.236 | 0.163 | -0.073 |
| Guerrero Vladimir | 0.196 | 0.119 | -0.077 |
| Huff Aubrey | 0.216 | 0.129 | -0.087 |
| Dunn Adam | 0.276 | 0.124 | -0.152 |
Asdrubal Cabrera so far has the highest jump and it is no surprise to to see that Adam Dunn is the leading decliner. The following is a look at the reasons behind the change for some of the hitters.
Droppers
Jayson Werth: After signing a $126 million dollar deal with the Nationals this last off season, Werth has been a massive disappointment. After 3 straight 5 WAR or better seasons, he is likely to finish 2011 under 3 WAR.
It should have been expected to see the 32 year old’s home run total drop since the 2010 HR park factor for Philadelphia was 120 and it is 100 for Washington (a 20% drop). He averaged 29 HRs over the last 3 seasons in Philly. Taking into account park factors, he should have been near 23 home runs for 2011. His 18 so far isn’t that far off.
To further illustrate this point, here is a chart of his home run and fly ball batted ball distances over the last three years:
2009 to 2011: 294 ft
2009: 293 ft
2010: 290 ft
2011: 301 ft
He is actually hitting the ball with as much authority or more this season compared to the previous 2 seasons.
One change, that is a little harder to explain, is the drop in doubles. It is a more difficult to hit a double in Washington (2B PF =94) than in Philly (2B PF = 99), but he looks to hit 1/3 fewer doubles this season (28) than last season (46). The cause for this drop is hard to pinpoint. The normal causes – a drop in line drive rate and slower running – aren’t happening in 2011.
Werth’s drop in ISO from home runs was somewhat to be expected because of moving to a larger park. The problem for him this season has been a drop in doubles, which is significantly harder to explain.
Vladimir Guerrero: Vlad’s drop off can be attributed to simply not hitting the ball as hard this season. His batted ball data lines up with his career numbers except for his 7.8% HR/FB%. It is down from last season’s value of 14.6% (career average = 16.1%). This drop is easily seen in his batted ball distances for fly balls and home runs.
2010: 293 ft average
2011: 273 ft average
Hitting the ball 20 less feet is not going to help a player’s power numbers.
Risers (I have already looked at the reasons behind Granderson’s and Cabera’s increases this season.)
Ian Kinsler: Two-thousand and eleven for Kinsler is not any kind of breakout season. Instead, he is just putting the disaster of 2010 behind him.
Here are his career ISOs since 2007:
2007: 0.178
2008: 0.199
2009: 0.235
2010: 0.125
2011: 0.232
Even though 2011’s ISO is a career high, it is just 7 points higher than the one he posted in 2009. The reason for the 2010 drop was that Kinsler was hurt for much of the season. He started the season on the DL with a sprained right ankle. Then on July 29th, he hurt his left groin (BTW – Am I the only one that doesn’t need to know which side of the groin the injury is on? Once I hear groin, I pretty much know the general area). He missed a total of 53 games over the entire season.
Now that he is healthy, the 29 year old is back to putting of similar number that he did before 2010.
Park factors are courtesy of Statcorner.com
Jeff, one of the authors of the fantasy baseball guide,The Process, writes for RotoGraphs, The Hardball Times, Rotowire, Baseball America, and BaseballHQ. He has been nominated for two SABR Analytics Research Award for Contemporary Analysis and won it in 2013 in tandem with Bill Petti. He has won four FSWA Awards including on for his Mining the News series. He's won Tout Wars three times, LABR twice, and got his first NFBC Main Event win in 2021. Follow him on Twitter @jeffwzimmerman.
Noting Jacoby Ellsbury’s ISO increase from 2009 (ignoring 2010) is a worthwhile addition.
I would have loved to see Ellsbury in here
Here’s Ellsbury’s last 4 years:
Year ISO
2008 .114
2009 .114
2010 .051 (only played in like 18 games)
2011 .216
I did a only HR piece a while back for the Roto side:
http://www.fangraphs.com/fantasy/index.php/ellsbury-is-the-home-run-spike-real/
But your piece completely ignores the most obvious factor : A completely changed swing.
Was Kinsler’s 2010 really a ‘disaster?’ Power/injury-wise sure, but dude had a .382 OBP (albeit aided by a .313 BABIP compared to .281 career).
I thought his 2010 represented a big change in his profile, to a BB machine with limited power (say, Barton 2010), vs. a big power/merely decent BB% guy in prior seasons (his ’08 looks a lot like Ellsbury 2011).
Now he’s got his power back, while sustaining his 2010 walk rate (~12% with a 20.5% O-Swing this yr), and is just missing the average. We may never see his ’08 again (.334 BABIP), but instead can we see a season where he puts all his skills together without the help of an overly generous BABIP, for .285/.385/.495 (Pedroia with more power)? That’s an MVP candidate obviously.
I was going to say the same thing. Kinsler keeps “adding to his game”, and IMO it’s just a matter of time before he puts it all otgether and combines it with some BABIP luck/fluke and has a monster season, something along the lines of 8+ WAR.
Right, as crazy as it sounds, I’d say .285/.385/495 could actually be conservative if his BABIP not only returns to decency, but pulls a James Shields and over-corrects itself from his .232 this year. If his secondary skills are real (why wouldn’t they be?), .315/.405/.535 with his defense and baserunning? Makes me blush.
Your Kinsler numbers don’t add up. You have 212 in the table at the top and 232 in the table in the bottom and neither is better than 2009’s 235.
I would prefer to see a player’s current year ISO compared to their career average, rather than to the previous season.
Yeah, this leads to things like Miguel Cabrera being on here because he had a huge, unsustainable SLG last year, and he’s slightly improved his BA this year.
Similar thing for V-Mart, his BA is up by 24 points over last year.
Hence the name “isolated” power
Isolated power is still BABIP dependant though, because BABIP luck doesn’t turn into all singles.
You take the same player, he’s going to have a higher ISO with a .400 BABIP than he is with a .200 BABIP
Miguel Cabrera has a “slightly improved BA” as in 0.001. Yep, that is a huge factor in this.
Pablo Sandoval also deserves to make the risers list with a .068 jump from .140 to .208. The data probably got compiled a few days ago because he just crossed the 400 PA mark on Tuesday night.
Sorry, I had been sitting on since Tuesday. I ran the data then. I will look at him later.
“It should have been expected to see the 32 year old’s home run total drop since the 2010 HR park factor for Philadelphia was 120 and it is 100 for Washington (a 20% drop). He averaged 29 HRs over the last 3 seasons in Philly. Taking into account park factors, he should have been near 23 home runs for 2011. His 18 so far isn’t that far off.”
– He only plays half his games at home so the decline would be 10% or near 26 HRs.
this
Obligatory annoying technical note: you’re not looking at the “highest and lowest ISO increases this season” but the highest ISO increases and decreases this season.
The “lowest ISO increases” would be players whose ISO is unchanged from last year.
If what you say is true, you can’t answer this question: which number is higher, -4 or -2?
Is there anybody who read the headline and didn’t imemdiately know that Adam Dunn would be at the bottom the list?
I didn’t expect his drop to be almost twice that of Huff’s. Huff had a great season followed by a terrible season. For Dunn’s drop in ISO to dwarf Huff’s is difficult to comprehend.
Until Kinsler can break a .720 ops on the road I can’t buy the chance he’ll have a monster year any time soon. Kinsler is the guy most consistently dependent on his home park I can remember.
You forgot about me (‘~’)
Or Me
But that’s actually reason for optimism. This year his BABIP is below .200 on the road. For his career it’s under .250. This year the ISOP is roughly equal home/road, so all of the difference in his OPS is from batting average. It’s true that in the past there was a wide variation in the power numbers.
But in addition to evening up the power numbers home/road this year, we’ve seen the SO rate decrease vs. 2009’s big power. His swinging strike rate is 3%! It’s known that Ian likes to make a lot of adjustments. If he ever decides to go with a consistent through the middle approach, that road BA will improve dramatically, and you’re talking Jose Bautista-lite – at 2B. That’s worth a few bucks.
I so called Kinsler this year to be awesome, was universally panned for my love of him
Awesome is when you can have a top 10 drop in iso and still be an all-star.
Among guys on the increaser list, I would bet on Upton as the one most likely to be close to this years level going forward the never several years.
ISO is treated in this article and others as if it was a stable power metric, not subject to the whims of BABIP. In reality, it’s pretty significantly correlated to batting average. It’s not like singles only randomly fall in, but doubles and triples are all power skill. As an example, if a guy has a .250 BA and a .400 SLG, his ISO is .150. Let’s say the next season, due solely to random variance, he hits .300. In order for his ISO to appear to be the same (.150), every single one of his additional hits would have to be a single.
The bottom line that if you want to measure raw power as a skill, it is much more reasonable to look at SLG/AVG than SLG-AVG.
var(iso) = var(slg – avg)
=var(slg) + var(avg) – 2 * cov(slg,avg)
The fact that they are correlated actually reduces the variance of iso, as you can see from the third term.