I did a search and found no discussions of how much difference it makes whether a player's speed rating is, say, 62, or 75.

I did a quick survey with a small sample and found that Chipper Jones 2008 at speed rating 57 batting leadoff scored once every 6.7 PA whereas Jim Edmonds 2004 at 75 speed batting 5th needed only 4.75 PA to score a run. Let's state that as 57-6.7 vs. 75-4.75. That's a big difference.

But at the same time, I found McGraw (1999) 86-5.42, Sosa (2001) 62-5.6; Bonds (2001) 69-6.05; Wheat (1924) 69-6.45. (Sosa's relative success might be due to context, as he's batting 4th on an offensive team with the most wins in its league.)

I'm guessing that the difference between 68 and 74, or 65 and 72, may be meaningful over the length of a season.

Any thoughts or information?

I have to think about it a bit more. I like the idea you brought up and how you came up with a measurement tool of sorts.

It appears that context may override speed alone. I just realized I should compare Wade Boggs with someone fast.

Boggs in 1988 scored 128 in 719 PA, slightly better than Lou Brock's 126 in 720 in 1971. Boggs is rated speed 57, Brock 92.

In WIS performance history, Boggs averages 121/641AB vs. Brock 127/668 AB.

My conclusion, Boggs' extremely high OBP trumps Brock's speed.

But what about John McGraw? Higher OBP and fast? His WIS performance average is 135 RS in only 509 AB.

Given similar OBP, it's clearly better to be fast than slow. How many runs might McGraw have scored leading off for the Bosox in 1988?

I know this is an old thread but did some digging based on performance history i took 4 guys with 90+ speed and 4 guys with ~50 speed and looked at their performance histories. All had 30+ seasons played. We took guys with similar OBP and ISO profiles so these guys all got on base but didnt hit too many XBH. I then looked at their non-home run times on base and how often those times got converted into runs. here is what we found;

Player	Season	SPD	Non-HR Onbase	Runs	Runs/non-HRS onbase
Butler	1990	94	271	119	0.439114
butler	1991	91	280	113	0.403571
Hamilton	1889	91	273	122	0.446886
rickey	1980	91	283	131	0.462898
AVG		91.75	276.75	121.25	0.438117
Tenney	1902	54	281	94	0.33452
Fain	1949	54	246	75	0.304878
Sheely	1924	54	268	84	0.313433
Boggs	1984	57	278	93	0.334532
AVG		54.75	268.25	86.5	0.321841

You can see speed converts nearly 44% of onbase to runs vs the slow 32%

We then looked at some median speed guys ~70 speed rating

Swanson	1933	77	262	107	0.408397
Pesky	2004	68	285	119	0.417544
Gillam	1956	77	276	110	0.398551
Rose	1973	72	296	123	0.415541
		73.5	279.75	114.75	0.410008

As you guys can see these guys convert closer to 41%.

We can then do an extra bit of math here trying to equate OBP -> Run Conversion. If our baseline is a .400 OBP 70 speed guy, here's how often the slow and the speed guy would need to get on base to get to our base line.

Summary Table

Speed	Conversion	Needed OBP to equal a .400 OBP 70-speed guy
50 (slow)	0.321	~.500 OBP
70 (median)	0.410	.400 OBP (baseline)
90 (fast)	0.438	~.377 OBP

TL;DR:
Speed helps, but the real difference comes from avoiding slow.

Going from 50 ? 70 speed gives a ~28% boost in converting base runners into runs — a huge jump.
Going from 70 ? 90 only adds another ~7% — a small gain.

So the return curve flattens fast:
?? 50 ? 70 = big jump
?? 70 ? 90 = marginal bump

Bottom line:
You don’t need burners — and if you roster plodders best not to have them as table setters.

thanks moonlight_do, great analysis.

another area where speed makes a huge impact (and this isn't in performance histories) is in GIDP. I think there you will see a big difference between 70 speed and 90+ speed. for example, in a regressive league I'm in, Rickey Henderson hit into 38 double plays...total over his entire career of 14,611 plate appearances. Craig Biggio meanwhile, who is pretty fast overall, had 106 in 12,850 PAs, so more than three times as likely.

this is why people love drafting those speedy 19th century catchers, with a team of all burners you're almost never going to hit into a double play.

If every single stat is exactly equal avg and ob% speed equal RS and RBI RC etc except slugging and one guy has 40 doubles and 10 triples and the other has 20 and 5 I’ll take the guy with all of the extra base hits.

That’s obvious. I’ll take that guy even if speed is 10 less. I think 10 difference makes a difference going 70 to 100. Under 70 is not a good lead off guy if other good alternatives.