Players Overrated by On Base Percentage
Victor Martinez leads the majors in wRC+. Victor Martinez doesn’t appear on the first page of our WAR leaderboard. These are the kinds of results that make people question whether defense is too heavily emphasized in WAR, and certainly, Martinez’s status as a full-time DH is the primary factor in the divergence between his offensive value and his overall value.
But I’m not trying to talk about that again. Instead, Martinez is instructive of another issue, as a reminder that on base percentage is not created equal for all players. After all, the entire goal of reaching base is to score runs, and players turn their baserunning opportunities into runs at different clips. In fact, the range of scoring runs as a percentage of times on base can be pretty large.
The league average player in the AL — we’re ignoring the NL because of pitcher hitting, for this case — scores 30% of the time they reach base on a non-HR. The guys who lead the league in turning times on base into runs are, as you’d expect, quite fast: James Jones leads the league at 48%, and not coincidentally, he’s also 25 for 26 in stolen base attempts this year. Jones’ speed allows his times on base to be more valuable than a player with the exact same OBP.
But then there are guys on the other side of the ledger. Victor Martinez isn’t the slowest guy in baseball, but he is a converted catcher in his thirties who has already had knee surgery, so not surprisingly, he doesn’t score as often when he gets on base. For the season, Martinez has come around on just 24% of his times on base, even though he’s had J.D. Martinez and his .538 slugging percentage behind him for most of the year.
Martinez isn’t last in the league in this category, though; In fact, he’s not even in the top 10. Here are the 10 players who have scored the least often on their run-scoring opportunities.
| Player | Run Scoring% |
|---|---|
| David Ortiz | 14% |
| Yunel Escobar | 17% |
| Adam Dunn | 17% |
| Dioner Navarro | 18% |
| Carlos Santana | 19% |
| Kendrys Morales | 19% |
| Mike Napoli | 20% |
| Kurt Suzuki | 21% |
| Jose Abreu | 22% |
| Kyle Seager | 23% |
Escobar is a surprising name, because he’s a shortstop with some speed, but this is probably just a function of the Rays offensive problems this year, as his career mark is 30%, and his career low before this season was 28%. Certainly, a player’s teammates offensive production can have an impact on this number, and Escobar is an example that this isn’t entirely within a player’s control.
But look at the rest of the names on the list. DH, DH, C, DH, DH, 1B/DH, C, 1B, 3B. This is basically a list of the slowest players in the game, and the guy who hits last in the string of decent hitters on the Mariners. Ortiz’s number is almost shockingly low, as he’s only scoring once every seven times he gets on base.
Speed has often been overrated in baseball circles, as it usually is found in guys who can’t hit all that well. But it does matter, and when evaluating offensive performance, we should remember that a nifty OBP from a baseclogger is not equal to that same OBP from someone who can actually run.
Dave is the Managing Editor of FanGraphs.
Ortiz and Napoli are obviously very slow, but they’re also above average hitters on a team that has been among the worst in the league on offense. The Red Sox have gotten absolutely nothing out of the second half of their line up all year.
I’m going to get behind this one. The overall Sox offense has been awful all year when you get past Napoli. I’m not the least bit surprised to see these figures.
Yes, and I also wonder how many times Ortiz in particular was taken off the bases for a pinch runner.
Whatever the case, this actually seems like the perfect opportunity to put a graph (or at least a table) together that shows how often a league average runner scores from first on a double or scores from second on a single (and possibly other situations) to get a sense for how much of these numbers are influenced by opportunity.
These guys are all big and slow, but they also (generally) hit in the middle of the order with lesser hitters coming up behind them. I would not at all be surprised to find the normal distribution of run scoring % to be 1-3 hitters most, 7-9 hitters second most and 4-6 hitters at the bottom, regardless of speed.
Intentional walks definitely something to consider here, especially in Ortiz’ case. Maybe if their on-base appearances were adjusted situationally, such as for how many outs and what base they reached.
Was batting order taken into account in this analysis?
Santana, Dunn, and Ortiz are also among the IsoD leaders, which, by definition, means they’re disproportionately trying to score all the way from 1st base.
You could also look at ISO leaders in reverse and the same would apply (this includes Escobar and Suzuki)
As in, reverse sort by ISO to find the most single-prone hitters who are reaching first by way of hit, not walk.
What an interesting concept – is there a way to do any hypothesis testing to compare players while assuming neutral lineup position and team? It would be interesting to see how much James Jones would score if he were on base in David Ortiz’s position.
It could be useful to just normalize by the hitting success of the batters following the player (by wOBA or similar). You could compare success vs. league average after getting on base using RE24 data (base/out states), but this would be tricky because the final base/out state is affected by both hitter success and player baserunning…
As others have done, I think this goes a little too far in the wrong direction. The whole point of OBP isn’t necessarily to score, it’s to NOT make an out. All these guys have control over is giving the next guy a chance. Base running is obviously a skill that players should be given credit for, but scoring is the wrong way to make the point. After all, if we only care about the one thing, why not just go all the way back to Runs Scored? It’s a step backwards. You are just illustrating a larger point here, but I think the point gets lost.
Seems to me this warrants more than a Instagraphs post. Any way to adjust for the rest of the lineup? Random variation? Historical career rates for players on the extreme ends of the spectrum? Effect on team run scoring?
It’s not just the speed but the instincts as well. Rendon has bested Span on the Nats on this metric despite having one fewer quality batter behind him because Span has run into more outs on the bases. Span likely has more raw speed on base but Rendon is apparently running the bases better. Perhaps this is why Rendon has also scored more runs than the Nats’ leadoff man.
Rendon keeps forcing Span out at 2nd 😉
For people like me who are also slow, it would be nice to breakdown how this relates to UBR. James Jones has a negative UBR, VMart has a very negative UBR. Wouldn’t something like % of times OB that you score runs be reflected in UBR?
Vindicated!
what about how often someone hits into double plays? I realize that is context related, and unlucky perhaps, but you don’t want someone with a ton of gidp, when a speedy runner can beat out more. of course if someone with speed can’t hit well, then that’s a problem.
What percentage of the value of OBP is in preventing an out, and what percentage is in score%?
Is the fact that a slow lead runner blocks speedier runners behind him from advancing factored into WAR in any way? Unless I am missing something, it simply hurts the faster runners.
Does negative value from baserunning runs properly adjust for this? That is: a batter earns a walk and is credited with X runs because a walk historically generates X runs on average. But because he’s slower than normal, walks from him are actually worth Y-amount less. Will this Y value be at matched at least somewhat closely by the negative value he’ll accumulate in the baserunning category?
Dave has a good point. Given this is an instantgraphs thing he probably didn’t have time to put this in. But his general hypothesis has some preliminary evidence. If you look at BSR. While Ortiz might be scoring less due to lack of lineup. One only needs to look at BSR to see that he’s not particularly good at advancing on the Paths. Now BSR takes into account all base running but this should be instructive since I view the value of getting on base as including how well someone legs out a double.
If you look at the link below sorted by BSR. You will see lots of names that are common with the above list and a almost everyone on this list as negative in BSR except for Escobar. I don’t think you can rely on this and the article to judge obp value for these players but it is true that these guys don’t score as often when getting on base due to their inability to advance as well as the average MLB player. Another reason why WAR is relevant. Many DHs tend to be hurt by the same problem its not merely defense and position adjustment.
Sorry here is the list
http://www.fangraphs.com/leaders.aspx?pos=all&stats=bat&lg=all&qual=y&type=8&season=2014&month=0&season1=2014&ind=0&team=0&rost=0&age=0&filter=&players=0&sort=18%2ca
What is the year to year correlation for individual player values? How subject to variation are these numbers? If Seager at 23% is a bottom 10 guy, that makes me think that most players tend to hover right around the 30% mean, even other slow players. It’s an interesting idea, but a few cases do not prove anything.