Are we going to miss the 80 stolen bases?

[DrungoHazewood]

BRob: 36 SB's in 2006, 50 SB's in 2007

CPat: 45 SB's in 2006, 37 SB's in 2007

If neither player is back in 2008, how badly will we miss that basestealing ability in the lineup?

We'll miss them. Without running the numbers I'd say the magnitude of that loss (assuming that they're replaced with zero steals/caught) is about 15 runs, or a win or two.

[Moose Milligan]

I'll miss watching them, definitely.

[mefogus]

We'll miss them. Without running the numbers I'd say the magnitude of that loss (assuming that they're replaced with zero steals/caught) is about 15 runs, or a win or two.

But your leaving out the decrease in Spunkitude that will occur at the loss of those SBs. They are directly correlated.

-m

[DrungoHazewood]

But your leaving out the decrease in Spunkitude that will occur at the loss of those SBs. They are directly correlated.

-m

True, but that's more than compensated for by the addition of Luis Hernandez. His yearly intangible contributions are the equivalent of Rickey Henderson's career SB total.

[High tops]

BRob: 36 SB's in 2006, 50 SB's in 2007

CPat: 45 SB's in 2006, 37 SB's in 2007

If neither player is back in 2008, how badly will we miss that basestealing ability in the lineup?

Don't worry Frobby, the way it looks, we'll have them both back, minus Miggy.

[mefogus]

True, but that's more than compensated for by the addition of Luis Hernandez. His yearly intangible contributions are the equivalent of Rickey Henderson's career SB total.

And by that measure, Hernandez should be worth about 30-33 wins next year.

-m

[Baltimoron]

Dan Fox calculates a baserunning runs total using: EqHAR (advancing bases on hits), EqGAR (advancing on ground outs), EqAAR (advancing on air outs), EqSBR (stolen bases, caught stealing and pickoffs) and EqOAR (which looks at wild pitches, passed balls, balks, and plays designated as "out advancing").

These metrics are derived from a multiyear run expectancy matrix and are based not only on the change in run expectancy created by these events but also the number and context of the opportunities the runners have. Its not perfect but it gives a good perspective.

Fox has BR as the 3rd best baserunner in 2007, at 8.43. Cpat is not among the top 15 in 2007.

Fox's best baserunners for 2007:

Name                 Opps  EqGAR  Opps EqSBR  Opps EqAAR  Opps EqHAR  Opps EqOAR EqRunsJuan Pierre            38   0.24    82  3.93    35  1.56    75  5.09   414  0.65 11.47Coco Crisp             36   1.11    35  1.95    30  2.48    47  1.64   323  1.41  8.60Brian Roberts          67   0.60    59  6.01    32 -0.36    56  2.58   402 -0.41  8.43Luis Castillo          46   0.95    22  0.63    45  1.00    61  3.07   602  2.78  8.42Jose Reyes             62   0.61   104  1.46    39 -0.31    46  3.57   393  2.56  7.90Jimmy Rollins          38  -0.01    48  3.81    52  0.83    51  3.33   425 -0.71  7.25Grady Sizemore         44   0.45    45 -0.19    43  1.70    54  4.01   435  1.22  7.20Johnny Damon           41   0.30    29  3.17    38  2.01    59  1.91   371 -0.32  7.06Dave Roberts           41   2.81    37  2.63    33 -0.15    40  1.81   281 -0.12  6.98Ichiro Suzuki          58  -0.45    45  2.52    54  0.54    73  2.35   507  1.77  6.72Chone Figgins          31   0.46    52  1.71    31  1.01    53  4.54   314 -1.10  6.63Curtis Granderson      43   0.58    28  4.07    56 -0.05    74  2.22   426 -0.56  6.26Jason Bartlett         40   0.62    28  2.81    18  0.21    52  3.51   325 -0.95  6.21Ian Kinsler            37   0.93    26  3.02    20  1.24    42  1.30   261 -0.30  6.19Kazuo Matsui           22  -0.59    36  3.89    31 -1.21    40  2.62   265  1.48  6.18--------------------------------------------------------------------------------------

BP link

From the same article, here are stolen bases, caught stealing and pickoffs. BR did very well, leading the major leagues:

Name               SBOpps  EqSBRBrian Roberts          59   6.01Curtis Granderson      28   4.07Juan Pierre            82   3.93Kazuo Matsui           36   3.89Jimmy Rollins          48   3.81Shane Victorino        43   3.76Rickie Weeks           28   3.47Johnny Damon           29   3.17Nathan McLouth         23   3.11Ian Kinsler            26   3.02

We know Corey Patterson's whole baserunning number was less than +6.18 runs, and likely less as Cpat was not top 10 in any of the 5 categories, and was actually 9th worst in EqAAR. Fox commented regarding Patterson's performance on air outs: "And then there are those like Corey Patterson, who didn't advance very frequently (just five times in 22 chances) and was also thrown out once trying to score from third, which cost him -0.60 runs."

Dan did post his entire baserunning spreadsheet on his website for 2006, so perhaps he will do the same for 2007. In 06 CPat was +3.85, and that did not include an EqOAR score.

[MP]

I tabulated some 2007 numbers to chew on. I think they're pretty compelling. I seperated the gamelogs of Roberts and Patterson into 2 buckets: games where they swiped at least one bag, and games where they didn't steal any. Look at the difference in runs scored.

Roberts games when he stole at least 1 base:

197 PA, 43 runs --> 1 run every 4.58 PA

Roberts games where he didn't steal a base:

519 PA, 60 runs --> 1 run every 8.65 PA

Patterson games when he stole at least 1 base:

138 PA, 28 runs --> 1 run every 4.92 PA

Patterson games where he didn't steal a base:

365 PA, 37 runs --> 1 run every 9.96 PA

So they score about twice as frequently in games where they're swiping bags. This doesn't seem negligible to me.

[Baltimoron]

I tabulated some 2007 numbers to chew on. I think they're pretty compelling. I seperated the gamelogs of Roberts and Patterson into 2 buckets: games where they swiped at least one bag, and games where they didn't steal any. Look at the difference in runs scored.

Roberts games when he stole at least 1 base:

197 PA, 43 runs --> 1 run every 4.58 PA

Roberts games where he didn't steal a base:

519 PA, 60 runs --> 1 run every 8.65 PA

Patterson games when he stole at least 1 base:

138 PA, 28 runs --> 1 run every 4.92 PA

Patterson games where he didn't steal a base:

365 PA, 37 runs --> 1 run every 9.96 PA

So they score about twice as frequently in games where they're swiping bags. This doesn't seem negligible to me.

You have a selective sampling issue because you have guaranteed that games where neither could score are excluded from the SB games, becuase you can't steal a base if you don't reach safely.

In other words, your categories are "games where we know they reached safely because they stole a base" and "games where we can't say if they reached but there are definitely many where we know they did not reach safely"

One of the biggest steps on the way to scoring is reaching base safely.

[Camden_yardbird]

I tabulated some 2007 numbers to chew on. I think they're pretty compelling. I seperated the gamelogs of Roberts and Patterson into 2 buckets: games where they swiped at least one bag, and games where they didn't steal any. Look at the difference in runs scored.

Roberts games when he stole at least 1 base:

197 PA, 43 runs --> 1 run every 4.58 PA

Roberts games where he didn't steal a base:

519 PA, 60 runs --> 1 run every 8.65 PA

Patterson games when he stole at least 1 base:

138 PA, 28 runs --> 1 run every 4.92 PA

Patterson games where he didn't steal a base:

365 PA, 37 runs --> 1 run every 9.96 PA

So they score about twice as frequently in games where they're swiping bags. This doesn't seem negligible to me.

Excellent statistics. I think that is pretty conclusive right there. The steals will be missed.

[MP]

You have a selective sampling issue because you have guaranteed that games where neither could score are excluded from the SB games, becuase you can't steal a base if you don't reach safely.

In other words, your categories are "games where we know they reached safely because they stole a base" and "games where we can't say if they reached but there are definitely many where we know they did not reach safely"

One of the biggest steps on the way to scoring is reaching base safely.

That's a good point right there.

I'll rework, thanks.

[MP]

Ok how about this. I'm not counting 3 games where Roberts' only non-out was a homer, and 1 game where his only non-out was a triple (since I'm trying to look at just the situations where he was in potential SB territory) The hopes of this is to see what affect Roberts' steals may have had on his scoring.

Roberts, games where he didn't reach:

22 games, 90 PA, 0 runs (naturally)

games where he reached and stole at least 1 base:

40 games, 197 PA, 43 runs --> 1 run per 4.58 PA

games where he reached but didn't steal a base:

90 games, 414 PA, 56 runs --> 1 run per 7.39 PA

You know, I'd like to say that's meaningful and builds a case for the affect of SB, but this is still too vague for me. There's nothing to say that he scored in the same inning as his SB, or that maybe he would have scored even without the SB. (for example he reaches 1st, swipes 2nd, then the batter homers) Then you get into non-quantifiable, proof-less regions like "did Roberts affect the way the batter was pitched" and such.

I'd have to look inning by inning through the game-logs to get really accurate numbers. And even then it could be debated. Ahh well...as Elwood Blues once said, **** this noise.

[mikezpen]

We didn't win withthe 80 steals, did we?

[mefogus]

You know, I'd like to say that's meaningful and builds a case for the affect of SB, but this is still too vague for me.

In other words, Patterson and Roberts score as frequently as the MLB average for the base that they steal and the number of outs on hand.

A useful equation would be:

((Total 2BSB - Total 2BCS) * Frequency of scoring from second) / 9

PLUS

((Total 3BSB - Total 3BCS) * Frequency of scoring from third) / 9

EQUALS

Expected wins generated from the SBs alone (totally independent of numbers of outs at the time of the steal).

This is essentially what Drungo was estimating up above. However, the equation above is overestimated because it does not take into account outs on hand at the time of the steal.

-m

[Frobby]

We didn't win withthe 80 steals, did we?

You can say that about absolutely anything. We didn't win with Nick Markakis, so should we trade him?

All I'm really trying to point out is that there are some aspects of the offense that stats like OPS don't capture at all, like team speed. If you had 9 Kevin Millars on your team, I'd bet the ratio of OPS to runs scored would be pretty bad.