The Brandon Phillips Paradox

[RedsManRick]

BP is not a mediocre hitter. That's a joke, right? He has 17-20 HR power, that alone sets him apart from the pack. From my understanding, .250 is now the mean batting average. so I see his .271 career avg as above average.

BP is fascinating statistically because he is moved around the line-up so much. Using the eye test, BP focuses on slugging and power when put in the clean-up spot. His approach changes when Dusty moves him to lead-off or the two-hole. The numbers don't show much difference but watching him play over the last few years does.

I was using mediocre as synonym for average. I realize it sometimes means below average; so sorry if that was confusing; my bad.

His career OPS+ is 96 (100 is average). The last 5 years: 103, 102, 118, 97,97.
His career wOBA is .327 (.330 is average). The last 5 years: .336, .336, .353. , .325, .316.

Simply as a hitter, he's almost the definition of average. Good contact rate, decent patience, decent pop.

Phillips is an above average 2B offensively. And he's a great defender. That makes him one of the best 2B in baseball. But we shouldn't conflate all of that with the reality that he's an average hitter who just so happens to be having a very "clutch" year in a situation in which he's been given a TON of opportunities to drive in runners.

Regarding him adapting his performance to his spot in the batting order, here's how he's performed:

[cost]
Split PA BA OBP SLG tOPS+ BB% K% ISO
Batting 1st 691 .265 .325 .426 100 6.4% 14.0% .161
Batting 2nd 819 .277 .320 .424 98 4.8% 12.0% .147
Batting 3rd 533 .276 .321 .425 98 5.3% 12.2% .149
Batting 4th 2400 .280 .330 .449 107 6.3% 13.8% .169[/cost]

Has he been at cleanup? Yep. A little bit. But let's not act like we're seeing wild swings in performance here. The differences are pretty much with the margin of error giving that we're talking about ~1 season in 1st/2nd/3rd and 4 seasons worth of cleaning up (that came during his typical peak years).

I think it's really important that we separate questions related to the value of a player's performance from the underlying talent driving that performance. that's where the RBI come in.

Phillips was similarly clutch last year in terms of hitting with runners on base. And yet, he only drove in 77 runs. This year he's only 3 behind that in 250 fewer PA. Should we be giving Phillips extra credit for the fact that his clutch hitting this year is substantially amplified by where Dusty put him in the order and the performance of the guys ahead of him? Not in my book. It's noteworthy that, from an RBI perspective, a high contact rate is actually really helpful. I wouldn't want/expect a patience & homers type like Dunn to drive in runs. Guys like Phillips who can hit for average but don't get on base a ton are actually pretty good "RBI producers".

As for all the other funny splits in the article in terms of home/away and so forth, if you dig hard enough you can find crazy splits like that all over the place. They don't actually mean anything. They just are.

[Trajinous]

Phillips is an above average 2B offensively. And he's a great defender. That makes him one of the best 2B in baseball.

Ha, you admit he's above average!

[RedsManRick]

Ha, you admit he's above average!

He's above the average second baseman. He's not above the average hitter. This distinction matters when constructing a lineup.

[Beltway]

Dat Dude has basically been a league average hitter during the first half of the season. The only paradox is how people can look at his other worldly situational performance this year and accept it as somehow something that is under his control when they'd flat out reject that notion concerning a similar stretch from some player on another team.

I was wondering the same thing. Allen Craig has similar types of splits (.274/.324/.760 with bases empty, .489/.500/1.189 with RISP) and I doubt anyone here would say Craig is a skilled situational hitter. They'd say his season is a fluke, he's getting lucky, and that'll he'll eventually normalize.

All of that aside, hitters tend to hit better with runners on base because the pitcher is working from the stretch. That accounts for some of the disparity, but obviously not all of it.

[WildcatFan]

Some people are overvaluing his offensive production this season simply because he's racking up RBIs. In reality, he's just getting more opportunities for them. He's on pace for almost 200 at-bats with RISP, as compared to around 140 per season the rest of his Reds career.

Not sure how much this brings to the discussion, but for reference's sake:

In his career, he strikes out about once every 6.5 at-bats.
In his career, he strikes out with RISP about once every 6.4 at-bats.
In 2013, he strikes out about once every 6.1 at-bats.
In 2013, he strikes out with RISP about once every 8.25 at-bats.

Not much to suggest he's a clutch hitter. Just that he's an average, and probably declining, hitter who happens to be coming up a lot with runners on compared to previous seasons (thank you, Joey and Choo).

[reds44]

He has been a much better hitter with RISP than with nobody on in his career, and it's a pretty large sample size.

[RedsManRick]

Ok. I do want to give Phillips due credit. He has definitely hit better with RISP over his career than he has hit with bases empty.

2006-2013		AVG	OBP	SLG	wOBA
Phillips Bases Empty	.266	.309	.432	.324
Phillips RISP		.287	.352	.437	.336
----------------------------------------------------
Phillips Split		.021	.043	.005	.012

But my understanding is that this is not uncommon; having runners on base helps hitters in general. So just how "special" is Phillips? I decided to look.

As a comparison group, I looked at all MLB hitters who had at least 1000 PA with RISP between 2006 and 2013. That produced a group of a comp group of 74 players. Here's how they fared:

2006-2013		AVG	OBP	SLG	wOBA
MLB AVG	 Bases Empty	.275	.341	.466	.352
MLB AVG RISP		.285	.379	.466	.355
----------------------------------------------------
MLB AVG Split		.010	.039	.000	.003

So, first let me recognize that this is a pretty select group of hitters. They had to be good enough to accrue 1000 PA with RISP over the past 6+ years. So this isn't all major leaguers. Of course, fewer PA than that and you're introduce more randomness. So I feel good about this group.

And Phillips comes out as an above average "clutch" hitters, measured as the difference between his wOBA with the bases empty and with RISP. Of those 75, Phillips would be tied for 22nd.

The list is actually pretty fascinating and close to normally distributed around that average of +.003. The top 3 guys are Youkilis (+.076), Rollins (+.064) and Berkman (+.046). The bottom 3 are Pedroia (-.032), Beltre (-.037) and Cano (-.054).

For the actual statisticians out there, here are the descriptive stats.
Mean .0033
Standard Error .0026
Median .0035
Mode .0100
Standard Deviation .0226
Sample Variance .0005
Kurtosis 1.092
Skewness 0.442
Range 0.13
Minimum -.0540
Maximum .0760

I'll have to think about it more to properly interpret it. But I think it's basically telling us that while Phillips has performed above average in terms of "clutchiness" -- raising his game with guys on base -- we cannot say with any certainty that Phillips is actually more talented at average (among proven MLB hitters) in terms of being "clutch". It could just be "luck".

[reds44]

That is fantastic stuff.

[Ironman92]

Ok. I do want to give Phillips due credit. He has definitely hit better with RISP over his career than he has hit with bases empty.

2006-2013AVGOBPSLGwOBA
Phillips Bases Empty.266.309.432.324
Phillips RISP.287.352.437.336
----------------------------------------------------
Phillips Split.021.043.005.012

But my understanding is that this is not uncommon; having runners on base helps hitters in general. So just how "special" is Phillips? I decided to look.

As a comparison group, I looked at all MLB hitters who had at least 1000 PA with RISP between 2006 and 2013. That produced a group of a comp group of 74 players. Here's how they fared:

2006-2013AVGOBPSLGwOBA
MLB AVG Bases Empty.275.341.466.352
MLB AVG RISP.285.379.466.336
----------------------------------------------------
MLB AVG Split.010.039.000.003

So, first let me recognize that this is a pretty select group of hitters. They had to be good enough to accrue 1000 PA with RISP over the past 6+ years. So this isn't all major leaguers. Of course, fewer PA than that and you're introduce more randomness. So I feel good about this group.

And Phillips comes out as an above average "clutch" hitters, measured as the difference between his wOBA with the bases empty and with RISP. Of those 75, Phillips would be tied for 22nd.

The list is actually pretty fascinating and close to normally distributed around that average of +.003. The top 3 guys are Youkilis (+.076), Rollins (+.064) and Berkman (+.046). The bottom 3 are Pedroia (-.032), Beltre (-.037) and Cano (-.054).

For the actual statisticians out there, here are the descriptive stats.
Mean.0033
Standard Error.0026
Median.0035
Mode.0100
Standard Deviation.0226
Sample Variance.0005
Kurtosis1.092
Skewness0.442
Range0.13
Minimum-.0540
Maximum.0760

I'll have to think about it more to properly interpret it. But I think it's basically telling us that while Phillips has performed above average in terms of "clutchiness" -- raising his game with guys on base -- we cannot say with any certainty that Phillips is actually more talented at average (among proven MLB hitters) in terms of being "clutch". It could just be "luck".

By my calculations "skewness" should be at 0.414

[gilpdawg]

Looking through bbref.com, by my count Phillips hit leadoff a grand total of 1 time while with Cleveland.

You beat me to it.

[Wonderful Monds]

Some people are overvaluing his offensive production this season simply because he's racking up RBIs. In reality, he's just getting more opportunities for them. He's on pace for almost 200 at-bats with RISP, as compared to around 140 per season the rest of his Reds career.

Not sure how much this brings to the discussion, but for reference's sake:

In his career, he strikes out about once every 6.5 at-bats.
In his career, he strikes out with RISP about once every 6.4 at-bats.
In 2013, he strikes out about once every 6.1 at-bats.
In 2013, he strikes out with RISP about once every 8.25 at-bats.

Not much to suggest he's a clutch hitter. Just that he's an average, and probably declining, hitter who happens to be coming up a lot with runners on compared to previous seasons (thank you, Joey and Choo).

Phillips' high RBI total is not just a function of having a lot of runners on when he hits. It's that combined with driving in an above average % of the runners that are on when he hits (23% vs the average of ~16%)

[membengal]

Those 2 excerpts reinforce what BP has been saying all year: he is trying to drive in runs at all costs. When nobody is on base or when he is way ahead in the count he is trying to hit the ball 9 miles (and he isn't good at it). He's excelling with RISP and/or with 2 strikes because his focus changes to putting the ball in play.

See, and here I was hoping he would focus on putting the ball in play and get on base all the time...