"It's going to take time. It's not going to happen overnight or in a week, it's going to take some time. I think we're all committed and we're all looking to do the best we can and to work on the things he [Gibson] wants us to work on. Certainly we want to cut down on the strikeouts."
-- Kelly Johnson
Through Tuesday night, the Arizona Diamondbacks had collectively combined to strikeout a total of 783 times. That's not just the most in the majors, it's 117 more whiffs than the second-placed outfit (the Marlins). It is also an 18% increase for our hitters on last season, when the team already led the majors. Mark Reynolds and Justin Upton are on pace to become the first team-mates to be #1 and #2 on the MLB strikeout list since 1965, and things appear to be getting worse.
Over the opening twenty games of the season, Arizona struck "only" 22.1% of the time - quotes used advisedly, since the 2010 league average is 18.9%. But in May/June, the K-rate increased to 24.4%, and in July, it has exploded even more. This month, an astonishing 31.2% of plate-appearances have ended with the Diamondbacks' batters taking their lumber back with them to the dugout.
What was a disease has now become an epidemic. After the jump, we'll look behind the raw K-numbers, to see if we can find out who are responsible for the plague, and whether there's any hope of a cure.
Firstly, who is responsible for the increase in strikeouts this season? Most obviously, it's not really Mark Reynolds - his increase of 1.2% is only one-third of the overall team number. Here are the strikeout rates for all the Diamondbacks with 50 or more PAs this season, and the comparable stat, where applicable, for the same player in 2009.
Part of the difference, is that the team has got rid of two of its least strikeout-prone players, and a third is hardly seeing any playing time. Whatever their flaws: Chad Tracy and Eric Byrnes did put the ball in play, with 2009 K-rates of 13.2% and 11.6%, and Augie Ojeda was even lower, at 9.1%. Between that trio, they had over 850 PA's at a very low strikeout pace of 11.2%. In contrast, none of the new arrivals - Abreu, LaRoche, Johnson, Ryal - are even at league-average, and so far have about the same number of plate-appearances, with more than twice the K-rate. The trade of Jackson also didn't help much, leading to increased playing-time for Gillespie and his 30% number.
On an individual level, however, the trend is for more strikeouts. Chris Young is the only one to have reduced their number by more than the 1%, matching the improvement of the other aspects of his game. On the other hand, a number of regulars have gone up by more than that. As well as Reynolds +1.2%, there's Drew (+2.0%), Snyder (+3.2%), LaRoche (+3.2%), Johnson (+5.8%) and, worst of all, Upton, all the way up at +7.4%. As noted above, he's now chasing Special K hard, to such an extent that we may have to start referring to him as K-Up.
The curious thing is, when you look at the plate discipline numbers, above, it's not like the team is swinging at bad pitches. The 2010 Diamondbacks swing at deliveries out of the zone 25.9%, the second-lowest number in the National League, and well below the median (28.6%). They're closer to average when it comes to pitches in the zone, but the overall number, regardless of location, still has only the Atlanta Braves swinging at less pitches.
Unfortunately, it's not a question of whether you swing or not that is the key factor in deciding the number of K's. I ran correlations between all the plate discipline stats and K-rate. Swing % had a correlation of -0.25: the more often you swing, the less often you strike out, but it's a fairly weak connection. However, Contact % - the percentage of time a team makes contact when they decide to swing - had a correlation of -0.91, which is extremely strong. Basically, the key factor in the number of strikeout by a team, is whether they make contact when they swing. And as the table above shows, that's an area where Arizona collectively blows chunks.
In or out of the zone, it doesn't matter: the Diamondbacks make less contact when they swing than any other team in the National League. But it is particularly pronounced out of the zone: the rest of the teams in the league range from 62.7% (Cincinnati) to 70.2% (Los Angeles) - Arizona are all the way down below 57%. The pattern is similar in the strike-zone: the rest of the league are from 85.9% to 90% (Rockies and Astros); the D-backs are, again, dead last by some way, at 83.8%. Combine these two, and the Diamondbacks' 75.2% contact rate is almost as far behind the 15th-ranked Marlins (78.7%), as the Marlins are behind the #1 Dodgers (82.3%).
Three individual Arizona players are present in the bottom eight for contact in the NL (and the top eight for K's). LaRoche is #8, at 73.3%, Justin Upton is #3 (70.1%), and Mark Reynolds "leads" the pack, all the way down at 61.6% - Adam Dunn is the only other player even below seventy percent, and barely, at 69.9%. Again, for Mark that isn't much of a change - his 2009 contact percentage was only 61.8%. For Upton, it's a little different. While his contact percentage is down somewhat, from 73.2% to 70.1%, he is swinging at many less balls in the strike-zone - down from 68.9% to 59.9%. This is something we looked at back in May, and much said then remains true.
The other part of the issue is when the strikeouts occur. As noted previously, if you strike out with two outs and nobody on, it's no different to any other out. But when runners get on base, it becomes more productive to move them over. You see this in general in the National League, because K-rate drops from 19.5% with the bases empty, to 18.2% with men on base, and all the way down to 16.6% with someone on third and less than two outs. The Diamondbacks hardly change their strikeout rate with the bases empty or not - it goes from 24.5% to 24.4%; it does then decrease to 21.3% with a man on 3rd and zero or one out, but is still significantly worse than average.
This ties, to some extent, into the statistics of "productive outs", which has been used as proof of Arizona's issues. However, that number has largely been discredited as having any significant effect on overall run production. While it has some inherent logic, the problem is not all outs that qualify e.g. sacrifice flies, bunts, and ground-outs that advance a runner, are truly "productive". A sacrifice fly with no outs and the bases loaded actually decreases the expected number of runs you'll score, from 2.42 to 1.97, so is hardly worth applause.
It also seems to become more of an issue later in the game. In the first three frames, the DIamondbacks strike out in 23.9% of plate-appearances. In the middle third, it's 23.1%. But in the seventh through ninth innings, the number goes up to 26.4%. And, again, it has been worse of late. Jack Magruder reports, "The D-backs have 22 strikeouts in the final three innings of their last three games. Asked about it, manager Kirk Gibson had a simple reason. "There is more pressure" in the late innings. "The game is harder to slow down," he said." There seems some truth to that, with overall league numbers also spiking later in the game, after dropping in the middle.
Can it be fixed? It really seems to come down to nothing more or less than getting the team to make more contact. It doesn't even necessarily have to be good contact, just enough of the ball to stop them from trudging sadly back to the dugout. I don't envy pitching coach Jack Howell's task in this area: as I type this, the team has just completed its franchise-worst fifth consecutive game with ten or more K's - if they do it again tomorrow night, they'll have tied the National League record. I don't know whether to cheer or boo them on in that quest...