Sunday, June 12, 2011

Tepid Optimism for Alex Avila

Alex Avila has been a more than pleasant surprise for the Detroit Tigers this season at catcher. After disappointing starts to the season from a number of Tigers, Alex Avila has done his best to keep the team afloat on offense. Through 204 plate appearances Avila has a .907 OPS with 9 home runs, 13 doubles, and even 3 triples. He currently has a .258 ISO and an absurd .389 wOBA both marks are first for qualified catchers by a mile (among unqualified catchers his ISO is second to Mike Napoli).

Needless to say, Avila has been one of, if not the top backstop in the MLB the first two months of the season. The question then becomes, can he keep it up? At first blush the answer would seem to be definitely not. This is the same guy coming off of a wOBA of .297 in his first full season and .112 ISO, numbers on the lower side even for a catcher. As a result, his improvement this season is well beyond any kind of reasonable expectation of improvement anybody could have had for him coming into this season.

Some of Avila’s peripheral stats have not shown any significant improvement. So far this season his K rate is 27% (K/AB) as opposed to 2010 when it was 24%, similarly his BB rate is 9.3% (BB/PA) as opposed to 2010 when it was 10.2%. Similarly Avila has almost an identical Line Drive rate this season. The one substantial difference this season for Avila is his ground ball and fly ball rates. This season he is hitting 34.3% GB and 44% FB as opposed to last season when he hit 43.5% GB and 35% FB. As we have seen generally ground balls help a players average while fly balls help a players power, unless of course a player hits a lot of home runs in which case the average actually will not suffer much from increased fly balls while power will increase. This season, Avila not only has hit more fly balls but has hit more of them out of the park with a 15% HR/FB rate as opposed to 10’ when he had a 9% HR/FB rate.

In 2010, Avila had a BABIP of .208, .101, and .696 for GB, FB and LD respectively. It should be noted that the FB and LD rates, both of which usually have very little differentiation among players, are unusually low. The mean rates would be in the range of .145 for FB and .725 for LD with a small standard deviation. As a result, as an initial matter it is possible Avila’s average last year was a bit flukey on the low end. Avila’s BABIP line this year is .255, .260, and .645 for GB, FB, and LD respectively. Obviously his GB and FB rates are pretty high and we can say almost definitely that his FB rate is not sustainable. However, he is likely to see his LD BABIP rise. Aside from BABIP the other noticeable increase is in Avila’s HR/FB rate, and in that case, we really do not have enough data on Avila in the majors to say what rate is probably most accurate. However a decent indicator is that thus far he has hit more home runs per plate appearance than he did in his best minor league season. While younger players often do discover power as they gain experience, Avila probably does not figure to continue such a notable increase in HR/FB. However, there is enough of a sample to conclude that it is possible Avila has become more of a FB hitter than he was last season and that his GB/FB rate might have shifted.

As a result, the best predication of Avila going forward would probably be to split the difference in HR/FB rate, and GB/FB rate while giving Avila credit for a LD rate during his better year and a FB rate closer to the league mean. So the final line is 21.5% LD rate with a .702 BABIP, 39% GB at .208, 39% FB at .145 with a 12% HR/FB rate. His line would look something like this assuming 450 AB.

450 x .73 = 329 AB

BA = (329 x .39 x .208 + 329 x .39 x .145 + 329 x .215 x .702 + 329 x .39 x .12)/450 = 110/450 = .244

HR = 329 x .39 x .12 = 15-16

As you can see assuming those numbers which are still fairly reasonable for Avila, his production suffers a steep drop. His final line would probably look something like .244/.332/.413 for a .745 OPS. That is a solid performance from a catcher but not exactly world beating.

The key to Avila’s continued production will be maintaining both a higher than expected FB rate and HR/FB rate. Avila’s average figures to decrease either way but there’s a big difference between 15 HR in 450 AB and something like 20+ HR in the same amount of AB.

Either way, Avila has too many positive trends to say he has not improved at all. So I am optimistic about Avila, but the remainder of the season will give a better indication of the degree to which Avila has improved.


  1. Great work, except the resulting prediction depresses me to no end.

  2. Well Actually I made a bit of a mistake for the end prediction. My prediction is basically a baseline for a whole season of production but it does not factor in his hot start which actually raises his expected mean production. So there's a very good chance his ending numbers are better than my prediction.

  3. Your calculation of Avila's batting average is incorrect.

    Given your projection that 27% of his at bats will end in a strike out, you conclude that 329 of his 450 at bats will result in a ball in play.

    Of these 39% will be flyballs, 39% will be groundballs, and 21.5% will be line drives. This adds up to 99.5% of all balls in play, just less than 100% due to rounding.

    So, using your projected BABIP with these figures the number of hits should be calculated as (329 x .39 x.208 + 329 x .39 x .145 + 329 x .215 x .702). Nevertheless, you include an additional calculation of (...329 x .39 x .12) which I can only deduce is the HR/FB rate. However that is already included in the equation as by definition his BABIP for fly balls (329*.39*.145)includes home runs. Even if this is not his HR/FB rate it cannot be included as you have already accounted for 100% of his 329 BIP.

    Using the correct equation his BABIP is 95/329 or .288 and his BA is 95/450 or .211

    Therefore there is even more reason to be pessimistic about his numbers for the rest of the season, especially if we assume his K/AB rate, which has increased this year, is stable.

  4. Dan,

    I'm aware about the totals being less than 100% but I'm fine with it as the remainder (bunts ect.) probably will not result in hits anyway and won't change the end result much.

    As for the BABIP calculation on fly balls, the BABIP only accounts for balls in the field of play and does not account for home runs. So the additional home run calculation would be appropriate. The mark of .145 is the expected percentage of balls that drop that both remain in play and drop in for hits while the .12 accounts for the home runs. Overall batting average for FB would be .265.