Yadier Molina’s clever situational adjustment

What bring us here today is a homer from Yadier Molina, one on August 8th. No, not his exciting grand slam a few nights ago, following the appearance of the rally cat. This one was just a solo shot. However, it was illustrative of a new approach Yadi was taking. Right before the homer, FS1 announcer Tom Verducci noted that Yadi utilizes a leg kick with no runners on, trying to hit for more power. With runners in scoring position, according to Verducci, he takes a more contact-oriented approach.

Now, I won’t speak on the mechanics of Yadi’s swing. That’s just not my specialty. I’ve had the good fortune of seeing a lot of games lately, but my attention isn’t typically focused on the players’ mechanics.

What I can do though, is find meaning in the stats. And from that, it sure seems that Molina is changing his approach based on the situation. We’ll look at two situations today: no runners on (226 plate appearances) and runners in scoring position (95 plate appearances). We’ll exclude situations where there’s just a runner on first (77 plate appearances), because the differences are more stark when doing so.

For starters, Molina’s ground ball rate with runners in scoring position (RISP) is 49.4%. With the bases empty, it’s 42.8%. Not only is he getting the ball in the air more with the bases empty, he’s getting better results. His home run per fly ball rate (HR/FB) with no runners on is 18%, compared to 7.4% with RISP.

That’s not all though. Let’s look at his results in both situations:

Molina has more swing and miss to his game with the bases empty, though 17.4% is still comfortably “below” the average strikeout rate. So far in 2017 though, he’s maintained a .200 ISO with no runners on.

With runners in scoring position, it appears he’s taking a more contact-oriented approach, with an extremely small K% that matches his BB%. For what it’s worth, the league’s strikeout rate also drops when going from no runners on to RISP, but it’s at a much less drastic rate (22.5% to 20.6%).

The batting average on balls in play (BABIP) is higher with runners in scoring position. This makes sense, as he would probably be more focused on hitting a hard grounder or line drive in those situations. But a whole season isn’t really enough to be very confident in a player’s BABIP, so we should be even less confident when chopping up 3/4th of a season into smaller sample sizes.

To get a better idea of Yadi’s contact quality, we’ll use the statcast data hosted at BaseballSavant.com. Specifically, we’ll utilize the statcast stats Exit Velocity (the speed of the ball off the bat) and Launch Angle (the vertical angle off the bat). To start things off, we’ll look at Yadi’s radial chart:

The protractor-shaped image above is used to represent any batted ball in terms of Exit Velocity and Launch Angle. Each of the dots is one of Yadi’s batted balls in 2017. The six shaded regions are the six qualities of contact. The types are listed on the right-hand side, and go from most productive to least productive.

This image is great at allowing one to visualize the different types of contact. However, it lacks context. Here’s a chart showing the league average performance of each quality of contact, the league average rate of each type, and Yadi’s rates in 2016 and 2017:

This spells things out much clearer. Molina hits more barrels with no runners on. Barrels are the very best batted ball, and are often a homer or a double off the wall. However, he also gets under the ball much more often. The “under” region is made up of pop-ups and medium velocity fly balls.

He hits a whole lot more Flares and Burners with RISP than when no one is on. Flares and Burners combine low velocity fly balls (bloopers or texas leaguers) with hard hit ground-balls (statcast classifies a grounder as below 10 degrees launch angle). Unlike Barrels and Solid Contact - which get value mostly from extra base hit potential - Flares and Burners are productive because they have a high likelihood of being a hit. Thus far in 2016, the league average BABIP on a flare or burner is .662.

With runners in scoring position, Yadi looks pretty similar to how he did last year, but better. One big difference is a much higher flare and burner rate, and a similarly lower “topped” rate. Topped batted balls are essentially weak grounders. His RISP approach seems awfully similar to his regular approach last year. With no runners on though, he’s trying to do damage in the air.

While his wRC+ is higher with no one on, xwOBA shows that he actually should have been much better with RISP (.369 vs .325 with no runners on). Granted, xwOBA doesn’t take his slow speed into account, but it still looks like the better approach so far. Perhaps as he gets used to this new approach, it’ll get better. After all, Yadi’s only going to get slower, and you can trot as slow as you want around the bases on a home run.

So we’ve established that Yadi is definitely taking a different strategy with no runners on as opposed to runners in scoring position. What we haven’t answered yet is, is it worth it? It makes sense that shortening up and trying to hit a single is better when say, there’s runners on second and third. It also makes sense that that same single leads to less value when there’s no runners on, especially with two outs, and especially when someone with Yadier Molina’s speed is the runner responsible for trying to score afterwards.

But hey, the fun thing about stats is challenging conventional wisdom, and that’s what we’re going to do. To do so, we’ll utilize the Run Expectancy chart found here. It lays out the expected runs scored the rest of the inning in each base-out state. That is, the expected runs scored when no one is on and no one is out, or first and third and one out, and so on. To save you a click, here’s is the full table:

Not really any surprises here. The more runners on, the farther along they are, and the less outs, the higher run expectancy. Using this table, we can find the increase in run value from a single, as compared to that produced by a homer, in each situation.

Using this article for rates on how often a runner attempts to score on a single, the chances he stays at third, and the chances he’s thrown out at home, we can find the average gain from a single in various situations. To simplify things, we didn’t include the chances that the hitter who singled attempted to go to second on a play at the plate or the chances that the ball is cut off, and the original hitter is caught going to second.

I also didn’t include situations where there is a runner on first, as we’d have to break it down into additional situations: those where the runner on first advances to third, those where the runner only advances to second, or those when the runner is thrown out attempting to advance to third.

We’ll also ignore the fact that runners are sent less often with less outs. The author of the linked article shares how often the runner scores from second on a single broken down by the number of outs, but not the chances he’s held or the chances of an out at the plate in those situations. So this is simplified a bit.

With all that in mind, I found the value of a single in a few different situations, as well as the value of a homer in each of those situations. Then, I found the value of a single, as a percentage of the value of a homer in each situation. Here are the results:

Molina appears to be on to something here. Bases empty represents three of the four situations where a single is least valuable, relative to a homer. The only exception is with a runner on second and two outs. The number of outs plays a factor because it means the chances of Yadi himself scoring afterwards become less likely.

However, runners are sent more often with two outs. The article linked above says runners score from second on a single 76% of the time with two outs, compared to just 39% with no outs and 51% of the time with one out. If I knew how often runners are held at third or thrown out at the plate, we’d probably find that the value of a single relative to a homer with a runner on second and two outs is higher than any of the three situations with no one on.

You might be inclined to think that a similar affect is in play with no outs, but that’s not really the case. Take the situation of a single with a runner on second and no outs. The run value of a runner on first and no outs is 0.859 runs and with first and third and no outs it’s 1.784. Add on the run that already scored in the first situation and it’s 1.859, just 4% higher than when the runner is held at first. That would go a long way to explaining the low score rate with no outs. Teams are willing to be more cautious with no outs, because the large majority of the time, that runner finds a way to score anyway.

There are also seems to be a connection here with two outs. In three of the four situations, the value of a single compared to a homer drops with two outs. Again, that’s because the chances of the original hitter eventually coming around to score becomes less likely. The only exception is a runner on third and two outs. That’s because the runner cannot score on a sac fly. The hitter has to get a hit in order to score the runner from third with two outs.

Usually, Molina earns his keep at the plate by being a singles hitter. He often has a very low strikeout rate, and maintains an above-average BABIP. He still has that profile in his back pocket, whenever he needs it. However, when singles are at their lowest relative value, he’s swinging for the fences now. The results aren’t bad, and the underlying numbers back that up.

Anyway, the point here is that Yadi’s adjustment seems to make sense. A single with no one on isn’t worthless, but it’s worth less compared to a homer than usual. Especially so when Yadi is the hitter, thanks to his slow speed. Add-on that Yadi is getting good results from this change in approach, and this seems like a great idea. It appears that this is just another way that Yadi pushes for every advantage the Cardinals can get.