Jerry Dipoto’s Nine Points of Offense

Among the comments made by Jerry Dipoto in his conversation with David Laurila at the recently concluded GM meetings, one finds a fascinating statement by the new Mariners general manager on the topic of identifying undervalued offense, and nine things that successful hitters do.

We’ve defined a nine-point criteria of what we believe a quality at-bat consists of. If you do those things, you can play here.

What a delicious quote. What could those nine points be?

I tried my hand at guessing in my last chat, and I drew upon the newest statistics and our knowledge of how batters age. I only came up with seven points, with some overlap — and some of this is not yet available publicly (if even privately).

1. O-Swing%
2. Z-Swing%
3. HR/FB%
4. swSTR%
5. Exit Velocity
6. Vertical Exit Angle
7. Horizontal Exit Angle / Oppo & Pull%

Theoretically, a batter who exhibited some ability by all these criteria would not swing at bad pitches, would swing at good pitches, would put his fly balls over the wall, would make good contact, would hit the ball hard, and would hit them all over the field with a line drive angle. You couldn’t defend the guy who was good at all of these things.

You’d have to weight these things correctly in order to cook up your secret sauce, and that might be hard given the lack of historical data with the last three entries, but it might be a fun way to get a jump on hitters. Which is difficult, considering how well-mined offensive statistics are.

Dipoto gives us some clues about what his list looks like. He talks about “productive at-bats” and says that “making an out on eight pitches is also a productive at-bat.” He also mentions bench coach Tim Bogar.

That last point is interesting. Bogar made a presentation at the Saber Seminar in 2012 about a metric called Offensive Value Percentage, or OVP. The idea behind OVP was to identify players who produce offensively beyond those way captured by readily available stats. He was thinking about outs that help the offense and the idea that more pitches is better.

Here were his eight components. Take note of number six in particular.

Eight Batting Components of OVP:
1. H
2. BB
3. HBP
4. Sac (bunt and fly)
5. Adv. the lead baserunner on an out
6. Eight or more pitch AB
7. GIDP/TP (Line out DP/TP only count as one, covered by defensive metrics)
8. First Pitch out of any inning

Do these traits form the first eight of the nine-point criteria cited by Dipoto? We can’t say, of course. But let’s pretend they do. What questions can we ask, and what answers can we provide?

A place to start our inquiry is with a point that struck a number of the conference’s attendees — namely, that all the variables in OVP are weighted equally.

Since we’re talking about looking at offensive production beyond the most commonly used metrics and the individual player, it makes sense to try and weight these events against win percentage on the team level. Do these various traits actually help teams win games?

Correlation Between OVP Stats and Team Win%

Statistic	Correlation to Win%	P-Value
H/PA	0.128	under .0001
BB/PA	0.124	under .0001
Pitches/PA	0.097	under .0001
HBP/PA	0.023	0.0032
SFH/PA	0.004	0.2069
GDP/PA	0.004	0.2299

P-values under .0001 suggest a relationship between the statistics beyond the effect of chance.
Correlations shown are r-squared.

This is a quick look, and it doesn’t have all of the components of OVP in it. But you can see quickly that not all of these stats are created equal in terms of driving wins.

If you evaluate players through this lens instead of more established ones, the top doesn’t look so strange. Jason Kipnis, Andrew McCutchen, Alex Gordon are the top three at least. Then there’s Adam Eaton fifth, and Micah Johnson tenth. Logan Forsythe is 16th. Atop the free agents list in this backwards-rigged OVP are the recently signed Franklin Gutierrez and the less-acclaimed Alex Avila, Mike Napoli, Ryan Raburn, and Drew Stubbs.

This ranking is only an approximation of OVP, though, so it’s not about those strange names, really. Surely, over time, the categories have been tweaked and weights have been found. We know there’s a ninth factor now.

And, really, even if these things haven’t demonstrated a great effect on team winning percentage, it’s an interesting way to try and find some value on offense, which has traditionally been seen as the aspect of the game that we are measuring best.

If you’re looking at what happens in an at-bat, there probably isn’t a revolution awaiting us on par with what on-base percentage offered in the early part last decade. At least until we have vertical and horizontal angles and exit velocities on all balls in play — in at least a two- or three-year sample — we don’t have any new public stats measuring the outcome of a plate appearance for the batter.

What this effort from Dipoto represents is a new attempt at incorporating traditional values into modern metrics. That is laudable, especially since his version probably looks different from the one we saw three years ago.

But it’s also important to remember that while pitches per plate appearance are strongly correlated with power and walks, they also add strikeouts and only have a small effect on weighted on-base average. Hitters also tend to see more pitches as they age, naturally, so a younger player who sees fewer pitches than an equivalent player is still probably preferable. Not all aggressive swingers are a drag on their team’s offense, either. So something that includes pitches seen per plate appearance as a metric threatens to overvalue the benefit of seeing that extra pitch.

Either way, it’s a fascinating look into the way the new Mariners front office might be run. Certainly not the way it used to be run.