On False Evidence

In an ESPN: The Magazine piece from this week, Amy Nelson and Peter Keating collaborated to report on allegations that the Toronto Blue Jays are stealing signs and signalling to batters via a mysterious man dressed in white who sits in right center field at Rogers Center.

While the evidence that the two writers present is mainly anecdotal, they also attempt to use statistics to point to Toronto’s supposedly outlandish success at home in 2010.

In 2010, the Jays swung at 48.9 percent of pitches, the highest rate in baseball. They hit just .269 on balls in play, the lowest in baseball by 12 points. However, they led the majors with 257 home runs, 46 more than the next-highest squad. In fact, the 2010 Jays had the highest isolated power (slugging percentage minus batting average) of any team since 1954. That’s what enabled Toronto to score 755 runs (ninth best in MLB) despite an abysmal team on-base percentage of .312 (fifth worst).

A huge proportion of the Jays’ power comes from their home ballpark. In 2010, Toronto blasted a whopping 146 homers at Rogers Centre, just seven homers shy of the all-time home record set by the Rangers in 2005.

Fair enough. Toronto played very well at home in 2010. But don’t teams in general play better at home than on the road? And is Toronto the only team that did this based on power alone? The article brings up isolated power numbers and attributes the Blue Jays’ offensive success to the team wide increase in power, by suggesting that “a huge proportion of the Jays’ power comes from their home ballpark.”

If we take a look at ISO home and away splits on a team by team basis we learn that Toronto has the highest rate at home and the second highest on the road. However, both the Rockies in the thin air of Colorado and the White Sox in Chicago have a larger positive difference in home and away splits than Toronto. In fact, the Blue Jays’ biggest accuser of fishy business, the New York Yankees, have a .046 difference in ISO between home and away, while the obvious cheats in Toronto have a .047 difference in ISO for the 2010 season.

The difference between home and away splits among the top five teams at home in terms of ISO:

  1. Toronto Blue Jays, +.047
  2. Colorado Rockies, +.074
  3. New York Yankees, +.046
  4. Arizona Diamondbacks, +.037
  5. Chicago White Sox, +.056

The article also suggests that individual players on the Blue Jays playing better at home than away is somehow evidence of foul play:

Several Jays had extreme splits in 2010. Bautista, for example, had a 1.118 OPS (on-base plus slugging) with 33 homers at home but an .879 OPS and 21 dingers on the road. First baseman Adam Lind had a .759 OPS with 15 homers in Toronto but a .660 OPS with eight bombs on the road. Second baseman Aaron Hill? His home-road OPS split was .730-.605. Shortstop Yunel Escobar was traded from Atlanta to Toronto in July 2010, and he has an .865 OPS at Rogers as a Jay but a .683 mark on the road. And then there’s Vernon Wells. The outfielder had a .990 OPS and 21 home runs in Toronto last season but crashed to .699 with 10 jacks away from Rogers Centre. This past winter he was traded to the Angels and has a .552 OPS in Halos home games.

Yeah. The home and away splits for Adam Lind and Aaron Hill are the biggest statistical anomaly for those two players in 2010, definitely not the .220 and .165 respective drop offs from the previous season.

We’ve already established that yes, the Blue Jays had more power at home than on the road, but that’s hardly anything unique when compared to other teams that relied on power hitting last season. Picking and choosing individual performances as signs of a massive home field advantage isn’t an honest sampling by any means. For ever Bautista and Escobar that performed better at home last year, there was an Edwin Encarnacion or John Buck who performed better on the road.

Wouldn’t it be more reasonable to look at the entire team’s OPS, and not just a select few? And from there, shouldn’t we look at how other clubs with good numbers at home last year performed on the road compared to the Blue Jays’ apparently staggering difference.

Let’s take a look at the difference between home and away splits among the top ten teams at home in terms of OPS:

  1. Colorado Rockies, +.212
  2. New York Yankees, +.090
  3. Boston Red Sox, +.027
  4. Texas Rangers, +.084
  5. Toronto Blue Jays, +.063
  6. Chicago White Sox, +.087
  7. Cincinnati Reds, +.042
  8. Arizona Diamondbacks, +.102
  9. Minnesota Twins, +.026
  10. Detroit Tigers, +.049

Of the top ten, five have a higher difference between home and away OPS than the Blue Jays, which to me suggests that if Toronto is stealing signs and signalling to batters from center field, they’re either not doing a very good job of it, or they could learn a thing or two from five other teams in baseball.

Even if you wanted to believe that signs from center field would make a difference for individual players, if he knew that it was coming, wouldn’t you expect Jose Bautista to hit more than two home runs off of change ups at Rogers Centre last year?

The real crux of the ESPN argument can be found in the work of Colin Wyers from Baseball Prospectus who finally compares Toronto to other teams and finds that:

Only the Blue Jays, and not their opponents, got a home run boost in Toronto. When the Jays were on the road in 2010, they hit home runs in 4 percent of plate appearances in which they made contact, compared with an AL average of 3.6 percent. At Rogers, their home run on contact rate soared to 5.4 percent, which is a home-field advantage seven times the magnitude teams typically enjoy.

What I’d be very interested in seeing is how other teams’ home runs on contact rates at home compare to their road split, which the article doesn’t offer. In 2010, the Blue Jays hit 67 more home runs at home than they gave up, which seems especially large considering that the next closest team was the Chicago White Sox who hit 34 more home runs at home than they gave up. However, this number appears to be a bit smaller when we see that Toronto also hit 40 more home runs than their pitchers gave up on the road, compared to those same White Sox who only hit nine more home runs on the road than they allowed. Again, Toronto emerges as having the largest split in home runs for versus home runs against at both home and away.

As a Toronto Blue Jays fan, I’m not so much of a homer as to blindly believe the defenses of Alex Anthopoulos:

That never happened, will never happen, not even a possibility. If it did happen, we’d be winning a lot more games at home. I think it’s a nonstory because no one ever has picked up the phone and called me about it. It’s never been an issue, and I would expect them to do so if it was.

Or Rogers Centre beneficiary Jose Bautista:

First of all, I don’t even know how you can do that. And second of all, it’s obviously something that’s not legal in the game. We do not cheat.

Both defenses sound somewhat similar to the “Who me? I wouldn’t even understand why you’d do that” explanation that Anthopoulos provided when rumours started circulating that his organization had a pre-draft agreement in place with first round pick Tyler Beede.

However, it’s pretty hard to use a single season’s worth of data to prove very much at all. It becomes even more difficult to take the supposed findings of an article seriously when the data that’s being provided is being manipulated with words to the degree that it is in this piece.

As we read through the anecdotes that begin the article, which range from the words of an angry reliever in the bullpen to broadcasters noticing the use of multiple signs, it reads to me as though the writers heard the anecdotes first and then went to the data to back up what was being suggested, which is always a dangerous way of going about interpretation, because it leaves a ton of data for another person to pick over (namely, me) and do the opposite.

Between the picking and choosing of numbers and the lack of comparisons to other teams, critically thinking about the piece ends up weakening their argument in a fashion I assume Nelson and Keating didn’t intend.

As for the anecdotal evidence, I can’t shake the belief that if stealing signs in this fashion was as wide spread at Rogers Centre as its being made out to be in this article, it would’ve gone to the commissioner’s office to be investigated long before it went to two ESPN reporters. The evidence being presented here isn’t a film crew getting caught filming an opposing team’s practice, it’s the meanderings of a few opposing players combined with picked over numbers being used to prove a point that doesn’t seem to exist.

That is the very opposite of convincing.

Generally speaking, if you summarize your article like this, you probably didn’t set out to do what you may have intended:

By themselves, these numbers are circumstantial evidence. Unsupported by data, the four players’ accounts might describe a scheme of uncertain impact. And without proper context, the Yankees’ decision to mask their signs could be chalked up to paranoia. But together, the numbers, the stories and the actions indicate one certainty: Every pitch to a Blue Jay in Toronto is worth watching.