Karl Ravech destroying Billy Hamilton on the BBTN podcast

[REDREAD]

Readred,

first of all, why use 2007 Run Scoring percentages? (I assume a typo?)

Second, is there really anyone that wants to base run scoring solely on OBP?

I'm going on vacation in a couple days, so I should have time to run the above lineup using Runs Created per PA. Should give us something more accurate.

I used the 2007 run scoring percentage because that was the first one that I could find.
I can't imagine it varies much from year to year, but if you want to find a more current one and recalculate, go ahead.
I have the link for the 30.1% in another post.

Many in sabermetrics believe that the most important thing to do is get on base. Right? And that RBIs are more or less a random event, not a skill. Therefore, it seems appropriate to use a league wide average scoring percentage on those extra theoretical 20 base runners that lineup juggling gives you.. Do you have a better method? Because even if you up it to 50% instead of 30%, that's still only 10 theoretical more runs over the season (not significant).

You do agree with the math that only 20 extra base runners are produced, right?

It seems shocking at first, but I remember doing the same math when we were arguing about Cozart batting 2nd, and it made little difference whether Cozart batted 2nd or 8th. The difference in number of at bats over the course of the season based on lineup position is relatively small, and it gets even smaller when you multiply by OBP, because there's honestly not that much difference between a .300 OBP guy and a .330 OBP guy. Sure, the 330 guy is better, but the impact of an extra .030 OBP gets way overstated in analysis.

[Raisor]

Baseball reference keeps track of the total PAs per season, so determine the percentage of runs scored per PA per season won't be difficult.

I can see doing that for simplicity, but it's going to be more accurate to use the actual RC/PA for the individual players.

Like I said, I'll look at it on my vacation.

[SteelSD]

You do agree with the math that only 20 extra base runners are produced, right?

No. You've excluded both additioal team PA gained by everyone else in the order; which excludes additional base runners as well as additional bases acquired. In short, ya' did it wrong.

[redssince75]

Plate appearances by batting order position(NL)

1st - 763
2nd - 746
3rd - 728
4th - 711
5th - 696
6th - 681
7th - 662
8th - 643
9th - 622

Hm. Since the plate appearances of slots 1-9 are in set order every game, simply dividing 162 by 9 should give you the expected difference in PAs between each slot. (That is, 1/9 of games are expected to end after the 1 hole hits, 1/9 after the 2 hole, etc etc). The answer is 18. A leadoff hitter (any leadoff hitter) should come up 18 more times per year than the 2nd place hitter. The 2nd place spot should have 18 more PAs than the 3rd. Etc.

Yet here are the differences

1-2 17
2-3 18
3-4 17
4-5 15
5-6 15
6-7 19
7-8 19
8-9 21


So what you can see -- basically, as expected -- is that with your better hitters bunched at spots 3-4-5, the game is relatively less likely to end after those hitting positions. And games are relatively more likely to end after the 7-8-9 hitters. Very interesting.

But on the whole, I don't think 18 more PAs per slot is anything to sneeze at. That seems to me to at least reach the verge of significance.

[REDREAD]

No. You've excluded both additioal team PA gained by everyone else in the order; which excludes additional base runners as well as additional bases acquired. In short, ya' did it wrong.

So you're saying since the Redszone lineup has 20 more baserunners, that means 20 less outs over the season..
Ok.. Let's assume a league average .320 OBP (non pitchers).. and give them 20 more plate appearances..
20 * .32 = 6 more baserunners.

Then those 6 extra baserunners *. 32 = 1.8 more

So maybe it's 28 extra baserunners over the course of the season.
That means 28 * .31 = 8.6 extra runs instead of 6 runs.
Still insignificant, and I'm being generous using .320 because the pitcher will consume some of those extra at bats.

If you are saying I did something else wrong, please recalculate it and show me.
I'm honestly interested in the truth here.
I just don't see lineups making that much of a real difference.

[wheels]

So you're saying since the Redszone lineup has 20 more baserunners, that means 20 less outs over the season..
Ok.. Let's assume a league average .320 OBP (non pitchers).. and give them 20 more plate appearances..
20 * .32 = 6 more baserunners.

Then those 6 extra baserunners *. 32 = 1.8 more

So maybe it's 28 extra baserunners over the course of the season.
That means 28 * .31 = 8.6 extra runs instead of 6 runs.
Still insignificant, and I'm being generous using .320 because the pitcher will consume some of those extra at bats.

If you are saying I did something else wrong, please recalculate it and show me.
I'm honestly interested in the truth here.
I just don't see lineups making that much of a real difference.

Those eight runs over the course of a season could mean a win or two. Those extra two, three wins mean a lot to a cuspy team like our Reds.

[swaisuc]

Those eight runs over the course of a season could mean a win or two. Those extra two, three wins mean a lot to a cuspy team like our Reds.

It could. Also, having Billy taking bases with nobody in front of him when he does get on could be worth 2 or 3 wins to a cuspy team like the Reds. It's not as easy to quantify a "right answer" as any of us want it to be. While I think it's interesting to quantify how many runs we may be sacrificing with BH leading off (maybe more than 8 if you factor OPS), I think the other side of the equation (how many runs we get because nobody is in front of BH) is much harder to quantify.

[bucksfan2]

Those eight runs over the course of a season could mean a win or two. Those extra two, three wins mean a lot to a cuspy team like our Reds.

Using hindsight Billy has created 2 runs this season that no one else in baseball does. The game against St. Louis and then last night in the 1st inning. Billy can do special things on the base path that no other player in baseball can do.

[SteelSD]

So you're saying since the Redszone lineup has 20 more baserunners, that means 20 less outs over the season..
Ok.. Let's assume a league average .320 OBP (non pitchers).. and give them 20 more plate appearances..
20 * .32 = 6 more baserunners.

Then those 6 extra baserunners *. 32 = 1.8 more

So maybe it's 28 extra baserunners over the course of the season.
That means 28 * .31 = 8.6 extra runs instead of 6 runs.
Still insignificant, and I'm being generous using .320 because the pitcher will consume some of those extra at bats.

If you are saying I did something else wrong, please recalculate it and show me.
I'm honestly interested in the truth here.
I just don't see lineups making that much of a real difference.

Probably the best thing you can do is google up one of the online lineup optimization tools. Even using some reasonable performance assumptions, the general "optimal" lineup would project to score around 4.3 to 4.4 Runs per game. That lineup would have Votto rather than Phillips in the leadoff slot BTW. The "worst" versions would project to score around 3.6 to 3.7 R/G. Over the course of a season, that's around 117 Runs due to the cumulative effect of adding PA over time for the optimal version coupled with the stacking of base-acquiition attributes.

Now 117 Runs is an extreme case; optimized versus crap so to speak. But we can fraction that and understand that half that effect is worth @6 wins; one-tenth almost two wins. Lineups do matter but, of course, matter less the longer a team sticks with an out machine in the leadoff slot (see: Taveras, Wily or Patterson, Corey). Right now Price should just move the whole lineup up a slot; with Hamilton helping by falling to 9th (or 8th for the faint of heart).

[bucksfan2]

Probably the best thing you can do is google up one of the online lineup optimization tools. Even using some reasonable performance assumptions, the general "optimal" lineup would project to score around 4.3 to 4.4 Runs per game. That lineup would have Votto rather than Phillips in the leadoff slot BTW. The "worst" versions would project to score around 3.6 to 3.7 R/G. Over the course of a season, that's around 117 Runs due to the cumulative effect of adding PA over time for the optimal version coupled with the stacking of base-acquiition attributes.

Now 117 Runs is an extreme case; optimized versus crap so to speak. But we can fraction that and understand that half that effect is worth @6 wins; one-tenth almost two wins. Lineups do matter but, of course, matter less the longer a team sticks with an out machine in the leadoff slot (see: Taveras, Wily or Patterson, Corey). Right now Price should just move the whole lineup up a slot; with Hamilton helping by falling to 9th (or 8th for the faint of heart).

Do those optimization tools factor in Billy Hamilton speed and base running? Guy can turn ground balls to the 1st baseman into hits. HBP to runs, BB's to 3b's.

I don't think moving Billy down in the order maximizes his value. He may struggle to get on at time, but he creates havoc when on. He isn't the type of guy you want up with runners on, he is the type of guy you want to be a table setter. Billy leading off an inning optimizes his skills vs Billy up with a runner on base, especially if a force out is created.

[Trajinous]

Do those optimization tools factor in Billy Hamilton speed and base running? Guy can turn ground balls to the 1st baseman into hits. HBP to runs, BB's to 3b's.

I don't think moving Billy down in the order maximizes his value. He may struggle to get on at time, but he creates havoc when on. He isn't the type of guy you want up with runners on, he is the type of guy you want to be a table setter. Billy leading off an inning optimizes his skills vs Billy up with a runner on base, especially if a force out is created.

Absolutely and well put.

[Always Red]

What are these numbers? Looks like whatever numbers they are, they are from last year also....Choo???????

While it's true that Choo is awesome, and we all love him, he is also on the very top of every chart that junkhead posts.

It is what it is. It all comes back to Choo.

[dougdirt]

to the "Redszone" lineup is only 20 base runners per year!

20 more base runners for Joey Votto and Jay Bruce may lead to 10-15 runs (of those 20 extra times on base, a few will be hits and a few will come with another runner on and drive that guy in, then Votto will drive in some of that 20 times on in and of those 20 times, Votto gets on base 8 times to add another runner on). That is pretty significant.

[REDREAD]

Those eight runs over the course of a season could mean a win or two. Those extra two, three wins mean a lot to a cuspy team like our Reds.

The Reds scored 698 runs last year. Hard to say if 8 more runs hyptothetically last year would've made a difference or not.
They say 10 runs is worth about 1 win on average.
So the difference is really about .8 wins.

Plus, those 8 runs are purely hypothetical, based on stats. It could be 10 more runs or zero runs.
Also the possibility that you are reducing the number of "Billy" runs by moving him down in the lineup.
The above analysis assumes Billy scores at the same rate as everyone else, which I don't think is true in reality.

Why sacrifice Billy H's development as leadoff hitter for 8 more hypothetical runs? Sorry, but it's such a small difference,
and theoretical, not worth making a major decision over.
Now maybe in a playoff game, if Billy H has OBP 250 all year, then you move him down to #7.
I can agree with that.. In the playoffs, you need to optimize everything possible.
But over the course of the regular season, it just doesn't matter.
Leake making a mistake and giving up that grand slam yesterday probably has a bigger impact on the season than whether Billy bats #1 or #7 (because the high probably is that's what cost us the game. There were other factors too, but that was the big blow).

[REDREAD]

Probably the best thing you can do is google up one of the online lineup optimization tools. Even using some reasonable performance assumptions, the general "optimal" lineup would project to score around 4.3 to 4.4 Runs per game. That lineup would have Votto rather than Phillips in the leadoff slot BTW. The "worst" versions would project to score around 3.6 to 3.7 R/G. Over the course of a season, that's around 117 Runs due to the cumulative effect of adding PA over time for the optimal version coupled with the stacking of base-acquiition attributes.

Now 117 Runs is an extreme case; optimized versus crap so to speak. But we can fraction that and understand that half that effect is worth @6 wins; one-tenth almost two wins. Lineups do matter but, of course, matter less the longer a team sticks with an out machine in the leadoff slot (see: Taveras, Wily or Patterson, Corey). Right now Price should just move the whole lineup up a slot; with Hamilton helping by falling to 9th (or 8th for the faint of heart).

I asked the thread for their optimal lineup and did the work.
If someone else wants to do an alternate analysis, they are welcome to.
I don't like lineup simulators if they are simple random number generators.
They just aren't realistic. We see things in baseball video games like the pitcher walking 7 guys in a row because the opposing lineup is full of high OBP guys. That just doesn't happen in real life.
Lots of assumptions in simulators.

Since this whole argument vs Billy H is OBP based, I think it's best to see how many base runners Billy leading off costs us.
The answer is 28 (unless I am shown to the contrary)

We can argue how many runs that means, but as a point of reference, last year the Reds had 1955 baserunners and scored 698 runs. That's a conversion rate of 35.7%.. So apply that to the 28 runners and you get 10 runs rounded up.. Still, not many.. 28 more guys on base out of 1955 (last year)..