The Diff: Reviewing stats on the surprising 2013 Indians

The Diff

The Diff is your weekly WFNY look into the amazing world of sports statistics. For a complete log of articles, click this link. Last week, I wrote on the varying playoff odds formulations. This week, I’m revisiting many of my stats comments about the Tribe in 2013.

The Diff

Today is Sept. 25 and the Cleveland Indians have over 80 percent odds of making the American League playoffs. Yes, that’s likely as one of two Wild Card teams, but all that matters is just making it into the postseason. Dating back to the advent of the Wild Card system in 1994, home teams are only 316-271 (.538) in the playoffs. That means it’s still very much an incredible toss-up in the probable one-game playoff that would take place one week from today, regardless of location.

Overall, the story of the never-quite-dead 2013 Cleveland Indians season could also be one for regression to the mean. This was a topic I first started writing about at the site back in mid-March in relation to standard MLB standings movement.

I looked at data from 1985-2010 to provide a solid sample size for this research. My conclusion: a team with less than 70 wins per 162 games has then improved by 9 wins per 162 games the next season. The Indians have already improved by 19 wins – nearing a franchise record.

Twice earlier this season – when the Indians were 26-18 and when they were 45-38 – I wrote about historical playoff odds for teams with that record. Both times, Cleveland’s odds were hovering around the 50 percent mark in the Wild Card era.

Now, because of a spectacular 16-6 start to the month of September, the Indians have defied those odds in the positive direction. As one final regular season edition of The Diff, let’s review even more items that I’ve discussed throughout the course of this incredible season.

The big free agent signings

One of my most popular editions of The Diff was called “Debunking myths about Michael Bourn” back on Feb. 13. Very few people predicted the team’s late signing of Bourn, despite the fact the Indians were one of a small handful of mentioned teams throughout the process.

On average from 2009-12, the speedy center fielder averaged 153 games and 5.0 WAR per season, per Baseball-Reference. This year, his WAR is only 2.0 in 126 games, representing the value of a purely average starter, not necessarily a $12 million/year star.

Some big factors are his below-average stolen base rates and relatively mediocre defensive evaluations. He’s never been a dynamic offensive threat. So while Bourn has been a disappointment thus far, my favorable comparison over Josh Hamilton still does hold true.

The most popular free agent acquisition, however, was Ohio State product Nick Swisher. He has 3.5 WAR this year after averaging just 2.9 WAR in 150 games per season from 2009-12. His recent hot streak and solid defensive play have led to one of his more surprisingly productive seasons.

Mark Reynolds, on the other hand, was an MVP candidate with a .291/.367/.645 line through May 9. Since then, he’s batting just .195/.287/.305 including his recent time with the New York Yankees. His season numbers aren’t too far off his career averages; he is what he is.

The Indians missed pretty substantially with the signings of Reynolds and Brett Myers. So in the end, the main free agents haven’t been a sensational factor in this year’s rapid improvement. More so, it’s been the rise of a number of surprising returning contributors.

Starting pitchers’ improvements

The internal improvement is most prominent within the starting pitching staff. Just as the season was starting, I wrote about the team’s awful quality start ratio back in 2012. The team truly had one of the worst staffs in recent American League history.

Likely nobody, not even new pitching coach Mickey Callaway, could have predicted the improvements that have taken place over the last half of the season for the Indians’ starters:

All 162 games in 2012: 5.25 ERA, 45.1% quality starts, 6.1 K/9, 3.5 BB/9, .284/.351/.451 line
First 87 games in 2013: 4.62 ERA, 41.4% quality starts, 8.0 K/9, 3.5 BB/9, .262/.335/.427 line
Next 70 games in 2013: 3.11 ERA, 50.0% quality starts, 9.0 K/9, 3.0 BB/9, .240/.304/.359 line

So yes, everything has improved: The ERA is down tremendously, strikeouts are way up, walks are down a bit and the opposition’s OPS has fallen drastically. The quality start rate is still near a league average 50%, as the team’s efficient starters actually haven’t been going the distance that often.

Last season, the downfall of the Indians was not only their abysmal starting pitching, but also their inability to compete when behind. The average AL team had a .313 winning percentage in non-quality starts. The Indians finished by far a league-worst 15-74 (.169) in such occurrences – including finishing the year on a 5-47 (.096) mark in this split. They had a slightly above average record of 53-20 (.726) in quality starts.

This season, they’re 31-55 (.360) without a quality start and 56-15 (.789) with one. That’s some incredible regression to the mean (thanks to improved starter efficiency, duh) in terms of overall record in non-quality starts. That .360 winning percentage would have ranked third-best in the AL last season.

Within this improved efficiency, the Indians rotation has set a number of recent franchise marks, a topic I first brought up in a stats headline in early August.

Ubaldo Jimenez and Justin Masterson each have 12 starts of five-plus innings and max one run allowed. This is the first pair of double-digit such performers for the Indians since 1974: Gaylord and Jim Perry.

All five of the Indians full-time starters have at least 22 starts and ERAs sub-4.15. The team last had five such starters in 1968: Steve Hargan, Sam McDowell, Sonny Siebert, Luis Tiant and Stan Williams.

Cleveland also has four starters (all but Zach McAllister) with 140 innings pitched and 8+ K/9. No other team in MLB history has ever accomplished this feat; Detroit, Washington and St. Louis all have three this year.

It’s been one hell of a ride. Heck, even the team wild pitches are down – after a historic start of 44 in 74 games to start the season (.595 per game), the Tribe only has 27 in the last 80 contests (.338 per game). That’s slightly below the established MLB average.

Mid-season offensive decline

Over the second half of the season, the story of the team also has included a surprising decline of the offensive production. It’s been serious enough that I dedicated an entire Aug. 22 version of The Diff to the bad offense.

Similar to the starting pitching, here’s a breakdown of what’s occurred from last season to the hot spring start and what has happened recently:

All 162 games in 2012: 4.12 runs/game, 6.9 XBH%, .251/.324/.381 line
First 46 games in 2013: 5.17 runs/game, 9.6 XBH%, .267/.336/.459 line
Next 111 games in 2013: 4.25 runs/game, 7.1 XBH%, .247/.322/.386 line

The offensive production from the last four months of May 24-Sept. 24 looks shockingly like the offensive production that Indians fans saw in 2012. Obviously, on the season, the 46-game outburst to begin the season certainly helped: The team still ranks sixth in the AL in runs/game and eighth in OPS.

So the Indians didn’t end up with the largest year-over-year slugging percentage jump in MLB history, as they once had back in May. Their offensive splits also aren’t that outrageous anymore: They rank just sixth in OPS with two outs, and rank third in OPS with two outs and RISP.

One of the bigger question marks that fans had entering the season was the team’s likely-to-be-increasing strikeout rate. With the additions of Bourn, Swisher, Reynolds and Drew Stubbs, there was no doubt it would jump up. Before the Bourn acquisition, I wrote about this topic in The Diff on Jan. 30: “Cleveland Indians embrace the strikeout.”

At the time, muddling through standard plate appearance distributions and career strikeout rates, I predicted that non-pitcher Indians would strikeout in 19.9% of their plate appearances. That wouldn’t lead the league, but it’d be higher than average. But again, I didn’t account for Bourn’s career 20.2% rate either.

Per FanGraphs, non-pitcher Indians have struck out in 20.9% of their 2013 plate appearances. That’s sixth-most frequently in baseball, still quite a bit behind Houston’s 25.3% mark. So yes, the prediction wasn’t that far off and again, strikeouts only are an excuse if teams aren’t producing otherwise: third-place Atlanta, sixth-place Cleveland and eighth-place Boston have most of the season.

A final note on close games

All of these stats lead me to this one final point: The 2013 Cleveland Indians are 29-17 (.630) in one-run games and 10-2 (.833) in extra-inning games. Those two records rank second-best and best in baseball this season.

Those are some extraordinary marks and represent an incredible turnaround from the 68-win Indians of one year ago. But alas, they’re also untrustworthy with one famous example staring Cleveland right in the face.

In 2012, the Baltimore Orioles shockingly improved from a 69-win club to a 93-win club. At the heart of this improvement: An MLB record 29-9 (.763) record in one-run games and similarly unfathomable 16-2 (.889) mark in extra-inning affairs.

This season, the Orioles regressed as many, many analysts expected. They currently are on pace for an 83-84-win season and were officially eliminated from playoff contention last night.

Entering the 2013 campaign, the WFNY staff predicted the Indians to win between 80-86 games. They’re already at 87-70 with five more games to play. I’d bargain to say that their largely unsustainable record in these close games has been the biggest contributing factor to the six-plus win boost we’ve seen compared to the preseason predictions.

So does that mean the Indians are doomed to regress in 2014? Perhaps, just like the Orioles. It’s certainly a crucial offseason ahead, as I wrote about the team’s outlook in early August. But for now, it’s been one hell of a ride. There are five games left and 80 percent playoff odds. We better savor this opportunity.

  • mgbode

    interestingly, we ‘likely’ will have 2 teams who finished 2012 with less than 70 wins make the playoffs this season (Boston being the other).

    even more interesting to me though is the difference in Pythagorean win totals between the Indians ’12 and ’13. We finished 2012 6 games better than our expected win total of 62. This season, we are only 1 game better than our Pythagorean expected total. That is an absurd jump (note: Only Houston, Miami, and Minnesota have expected win totals at or below 62 this season. So, imagine one of them making the playoffs next year.).

  • mgbode

    I would argue that their regression in close games (likely) may be offset by an increase in offensive output (as too many players on offense are having very down years compared to career averages).