Cobb vs Ruth

[Comm]

How can you rate Cobb as a better hitter than Ruth? Ruth had way more RBI's and averaged more Runs in a season then what Cobb did, and it's Runs that win the game, also.

[StresManagement]

Cobb was nowhere near the hitter ruth was. A better comparison to ruth as a hitter is ted williams or barry bonds ( if you ignore the balco scandal).

[Sultan_1895-1948]

How can you rate Cobb as a better hitter than Ruth? Ruth had way more RBI's and averaged more Runs in a season then what Cobb did, and it's Runs that win the game, also.


I attempted to find a "opening can of worms" smiley, but there wasn't one. This topic has been hashed through before. Granted, pretty sure that was discussing them as ballplayers, and not just hitters.

I'm as big of a Cobb fan as there is on here, outside of Chris elder, Bill, and possibly a couple unknowns, but as pure hitters, there's not much of a comparison imo. I don't think Cobb's initial gift was natural hitting ability. He had the skills to hit, and later honed those skills to fit what he'd learned through strategy and analyzing, but he relied on his brains and legs much more than raw hitting ability. When it all came together, his approach was genius. I think more than any other player, he got more out of his ability; that is, rising further above his expected production than anyone before or since.

Babe was the package. Unselfish enough to take what was given to him, and keep opposing defenses from using a shift, even bunting on occasion. Talented enough to be among the BA leaders while taking his overall approach. Smart enough to not expand the strikezone too far. Skilled enough to strike out seldom, relative to other sluggers throughout history. I think his power swing was tailored before he ever got to the bigs, but that his early years as pitcher allowed him to master a high avg/power swing against major league pitching.

[Edgartohof]

How can you rate Cobb as a better hitter than Ruth? Ruth had way more RBI's and averaged more Runs in a season then what Cobb did, and it's Runs that win the game, also.

I'm not saying that Cobb is a better hitter, but just to play devils advocate, I'll give it a quick shot here.

First off, he has the Major league career BA record of .366, beating out 2nd place by 8 points, and had this average over 11,000+ AB's! He won 11 BA titles (5 consecutive), and was 2nd four other times, not including all of his other top 5 and top 10 appearances. He hit over .400 twice, including a .420 BA in 1911.

He also led his league in OBP and Slg. quite frequently, with 7 and 8 titles respectively. He was a hit magnet, doing whatever it took to get the hit, and over his career, became the all-time leader in hits (until Rose). He also had over 200 hits/season 9 times, with over 220 hits/season three times (with a career high of 248).

He also had good doubles and triples power, ending with over 700 doubles and nearly 300 triples, and leading or near the top in both of these categories quite frequently. He did not have the HR as a tool to use, for most of his career, and was already more than halfway through his career when the liveball era came about, and if he wanted to hit HR's (and he could have if he had wished to do so), he would have had to change his style, to change what had worked for him for his whole career, what had made him the best of the best for so long, and even without HR's, he was still great.

He was dominant, and he lasted a long time, playing 24 seasons, and ended his career with many career records, some of which he still holds, or is still near the top, such as: BA, OBP, runs, hits, RC, and many others.

And as was already mentioned, rarely is a case made that Cobb was a better hitter than Ruth, as hitting wasn't his only asset (and maybe not even his best), but is better compared as a total ballplayer, which he was.

[Sultan_1895-1948]

he would have had to change his style, to change what had worked for him for his whole career, what had made him the best of the best for so long, and even without HR's, he was still great.


Solid post Edgar HOF supporter. Couldn't agree with this more. For Cobb to abandon his chosen style, would be like a successful artist going from 9 colors on his pallet, down to 3. Or something like that. What made his place-hitting so valuable was his speed, aggressiveness, and his preparation/strategy. To bail on the contact/place hitting, would be to fly out far more often and strike out more, disallowing him to use his other gifts as much as possible. And shouldn't that always be the goal. Conform your approach to best use your assets? Cecil fielder would never be a slap hitter, and Gwynn never changed to a power approach, despite seeing how the game begged for it, although I heard after talking to Ballgame he made a slight adjustment on the inside pitch.

[digglahhh]

I'm in Ruth's corner, but just a quick point here.

You know the old baseball adage, "A walk is as good as a hit."

Well, in Cobb's case, a walk was as good as an extra base hit. His SLG or OPS doesn't take into effect that one pitch after a single, or walk he's on second.:)

I hear the argument that Cobb's power was based on his speed and if you want to make the argument that because of that he wouldn't have hit a lot of homers in today's game (as you might think he would b/c he led his league in SLG) that's okay. But on its own terms, the argument does nothing to knock Cobb's power, for a player to hit a triple, all he needs to do is to hit the ball as a far a HE needs to hit it to get to third. Perhaps Cobb didn't have to hit it as far as others, but that's to his credit.

[538280]

I'm in Ruth's corner, but just a quick point here.

You know the old baseball adage, "A walk is as good as a hit."

Well, in Cobb's case, a walk was as good as an extra base hit. His SLG or OPS doesn't take into effect that one pitch after a single, or walk he's on second.:)


Sure, he was on second about 50 times, but he was also off base about 20 (caught stealing). The extra base is nice, but being off base IMO is a bigger penalty that the gain for reaching second is. Pete Palmer in The Hidden Game of Baseball does a rigourous study of game accounts to place run values on offensive events, and he finds a SB increases run scoring by 0.2, but a CS decreases it by 0.35. By stealing 50 bases a year and being caught 20 times (which was probably about Cobb's average based on CS data we have), he was only contributing about 3 runs.

[SABR Matt]

Careful.

In the deadball era, being on first was not very helpful compared to being on second in terms of run scoring. Because most offenses scored by stringing together singles, the run expectation for a runner on first was probably much lower than it was for a runner on second.

50 steals against 20 CS is probably a very valuable thing in the deadball era.

[Taco De Muerte]

Cobb shouldn't be rated ahead of ruth as a hitter. He wasn't the hitter ruth was. Williams, Bonds, and even mantle are better comparison's - Mantle at his peak was probably just as good if you make a league quality adjustment.

[Ubiquitous]

50 steals against 20 CS is probably a very valuable thing in the deadball era.

I doubt it. 20 cs is 20 outs which is 20 potential stoppages of the string. The 71% success rate was/is probably no more valuable then it is now. Which is probably just above average.

But of course I disagree with that whole notion of success rates for stealing bases anyway.

[digglahhh]

I doubt it. 20 cs is 20 outs which is 20 potential stoppages of the string. The 71% success rate was/is probably no more valuable then it is now. Which is probably just above average.

But of course I disagree with that whole notion of success rates for stealing bases anyway.

If XBHs are less common, the the importance of a SB is increased and the detriment of a CS is minimized.

Isn't that the underlying philosophy of the SABR line-up models that call for Jose Reyes to hit 6th?

I don't know exactly what you mean by the last part of the post, but if you mean what I think you do, I agree. Stolen bases are perhaps the best example of the notion that general sabermetric truths can be completely contradicted in any individual situation without being disproved in the general sense. With ther notion that a CS is more detrimental than a SB is beneficial, I think that SB situations should be considered far more individually and specifically than they are in the majority of SABR analysis I've come across.

[SABR Matt]

I certainly don't disagree with the idea that perhaps our analysis of baserunning is giving too much positive credit for steals and fast baserunners in many instances...

I do think though that a SB in Cobb's day was worth a lot more than it is today.

[Ubiquitous]

If XBHs are less common, the the importance of a SB is increased and the detriment of a CS is minimized.



No not really. If XBH are less common (and hits and walks in general) then one needs more events to score so while the importance of stealing a base might increase so does the importance of not getting caught. Throw in errors and the importance of not removing a base-runner becomes even more important.


As for stealing bases I doubt they worth a lot more.

In 1916 the average NL team scored 3.45 runs. If the run value of a steal went up a lot it would mean that a single steal would be worth more then 10% of the total runs scored in that game for a team.

Now perhaps you are saying that a steal in terms of a % is a lot more. In that its raw run value might be less then today but in terms of impact it is far greater. To that I still doubt. Teams were averaging either a little under or a little over a steal per game. Now it is possible that more stolen bases of home took place which would create an illusion that a steal is worth mor. Since higher leverage stolen bases were happening. But in terms of a runner stealing second I doubt the value changed much. I doubt the value of a steal of third changed much, and I severely doubt that a steal of home changed much at all. Its just that the thre frequency of the three changed thus creating a different (though I doubt it changed a lot). Sore of like Diggs view on Coors field and away splits.

A player stealing second base once a game just isn't going to be that impactful. Right now I believe the overall steal is credited with being worth around 5 to 7% I can't see its worth increasing by a lot? Afterall what is a lot? To me moving from 7% to 8% isn't a lot.

[Sultan_1895-1948]

No not really. If XBH are less common (and hits and walks in general) then one needs more events to score so while the importance of stealing a base might increase so does the importance of not getting caught. Throw in errors and the importance of not removing a base-runner becomes even more important.


I tend to take the other side of this. When a player reached first base, it would usually take at least 2 more hits to get him home. While stealing was a gamble, there was much more reward than risk back then. Getting into scoring position meant just one single to get you home. You're basically taking the 65% chance of stealing, as opposed to waiting for two events that come in at a 30% chance each.

How did a Cobb vs. Ruth hitting discussion turn into a SB discussion anyway?

[digglahhh]

Ubiq,

I see where you are coming from, especially on the leveraged steal hypothesis. But, I think this comes back to what I was saying about SBs being events that are best suited to be studied on an individual, situation-specific basis.

With a runner on first base and singles hitters up, a successful SB is valuabe becase now a runner can score on a single. If XBXs, that can score a runner from first, are less frequent then it stands to reason, that a runner can get thrown out more often and be redeemed because he wouldn't have scored anyway. Now, that extra out that he creates by being thrown out, depriving another batter an opportunity, causes us to never really know for sure if he would have scored had he not been thrown out, and that's a problem.

A successful steal of second highly increases the likelihood that the runner scores on one hit as opposed to two. Fewer overall runs would put a premium on each run, meaning that all opportunities become higher stakes and both successes and failures become more profound. I could see either side of the argument theoretically, but I think we'd need to see the actual data to bear it out.

[Brownie31]

Isn't comparing Cobb and Ruth made difficult due to their
radically different styles?

The same applies to comparing sportswriters of that era.
There were the more idealistic "gee whiz" Grantland Rice
types and the cynical "Aw Nuts!" Westbrook Pegler types.

Brownie31

[csh19792001]

If XBHs are less common, the the importance of a SB is increased and the detriment of a CS is minimized.

Isn't that the underlying philosophy of the SABR line-up models that call for Jose Reyes to hit 6th?

I don't know exactly what you mean by the last part of the post, but if you mean what I think you do, I agree. Stolen bases are perhaps the best example of the notion that general sabermetric truths can be completely contradicted in any individual situation without being disproved in the general sense. With ther notion that a CS is more detrimental than a SB is beneficial, I think that SB situations should be considered far more individually and specifically than they are in the majority of SABR analysis I've come across.

Exactly. I'm with Matt, Sultan (and perhaps Derek) here.

And it isn't just coincidence that most of the greatest basestealers before roughly 1920 usually had (by today's standards) lousy SB percentages.

It was almost a completely different game, and with the paucity of extra base hits, the basic non existence of homeruns, and the shoddy gloves/equipment (leading to tons of errors, bad bounces, likelihood of passed balls, etc), it was MUCH more worth the risk to steal/take the extra base in the deadball era than it is today. Again, we have people imposing the modern "Moneyball" paradigm onto players of 100 years ago. This is misguided and once again devoid of historical perspective.

[Ubiquitous]

Ubiq,

I see where you are coming from, especially on the leveraged steal hypothesis. But, I think this comes back to what I was saying about SBs being events that are best suited to be studied on an individual, situation-specific basis.

With a runner on first base and singles hitters up, a successful SB is valuabe becase now a runner can score on a single. If XBXs, that can score a runner from first, are less frequent then it stands to reason, that a runner can get thrown out more often and be redeemed because he wouldn't have scored anyway. Now, that extra out that he creates by being thrown out, depriving another batter an opportunity, causes us to never really know for sure if he would have scored had he not been thrown out, and that's a problem.

A successful steal of second highly increases the likelihood that the runner scores on one hit as opposed to two. Fewer overall runs would put a premium on each run, meaning that all opportunities become higher stakes and both successes and failures become more profound. I could see either side of the argument theoretically, but I think we'd need to see the actual data to bear it out.

Unfortunately Tango's custom linear weights only go back to 1919 but what I see in that data is that the SB and CS stay about the same throughout history. Which means that both do in fact become more important because they have the same run value but the teams overall are scoring less so each is now worth a greater % of the teams run scoring average. For the most part it looks as though SB-CS follows the 66% average that we have come to know (and disagree with). Usually a CS costs a team 10% of its run total while a SB counts toward 5% of the teams runs.


Now then everyone keeps saying that a successful steal shortens the chain of events to scoring. Yes that is true but that doesn't mean that a caught stealing is now worth less. A CS lengthens the chain of events need to score a run. Now instead of 2 hits in 3 outs the team needs 3 hits in 2 outs to score a run. Something that it is not likely to do.

[Ubiquitous]

Exactly. I'm with Matt, Sultan (and perhaps Derek) here.

And it isn't just coincidence that most of the greatest basestealers before roughly 1920 usually had (by today's standards) lousy SB percentages.

It was almost a completely different game, and with the paucity of extra base hits, the basic non existence of homeruns, and the shoddy gloves/equipment (leading to tons of errors, bad bounces, likelihood of passed balls, etc), it was MUCH more worth the risk to steal/take the extra base in the deadball era than it is today. Again, we have people imposing the modern "Moneyball" paradigm onto players of 100 years ago. This is misguided and once again devoid of historical perspective.


Geez there is no great conspiracy here, nobody is imposing anything. Nobody said that a stolen based or trying to steal wasn't worth more. What I am saying is that a caught stealing meant more as well. For exactly the same reasons that a stolen base was worth more. If you get caught stealing a base you are no longer a baserunner and your team receives an out. You now can no longer take an extra base on an error, you can take an extra base because of the field or the gloves or the strategy of the day, your gone.

Secondly its odd that a poor success rate somehows shows you that it was better to run in those day. That means the opposite. That these players should have picked and chose their spots to steal better. Being right half the time on base is not a good success rate. Being right less then half the time is not good either. If you are standing on first base 100 times and you have a 10% chance of scoring and by not making an out your fellow batters have a 5% chance of scoring that means on average 15 runs will score. Every time on first you try to steal second. You do it successfully 50% of the time and standing on second you score 15% of the time and the batters still score 5% of the time, you now average 10 runs a game (well actually a little higher since batters would still have a scoring % even with you caught stealing).

Every kind of player was trying to steal a base and a lot of them shouldn't have been trying. Saying that isn't devoid of historical perspective ignoring that is though.

[Ubiquitous]

Also I'll say it again the players back then were attempting different kinds of steals. Meaning it wasn't the same ratio of steals today. 85% steal of second, 14% steal of third, and 1% of home (just made those up). They were trying to steal third and home more often.

Retrosheet has the incomplete 1911 NL season and so far they have 691 stolen bases and 826 CS. The Giants had over 200 CS alone and in looking at their dailies there was a lot of stupid baserunning. Failed attempts at stealing third with no outs, failed attempts at stealing home. Just a lot of bad decisions. Players were running until they were thrown out or scored. Yes the scoring part is a positive but the out part is a negative.

Someone who can better play with databases could probably present the data better then I can, and show us exactly what kind of steals were being attempt in 1911. Whether they stole it or got caught in what kind of situation and then also show us a modern seasons stolen base numbers as well. I would love to do it but it would probably take me all day to get the 1911 part. I could probably do a modern season easily though.

[misterdirt]

For what its worth, I believe that Cobb had a greater ratio of runs/(hits+walks) than any of the players of his era (in his prime years). If his steals minus caught stealing were not a scoring benefit to his team that would not have been so.

[digglahhh]

The Giants had over 200 CS alone and in looking at their dailies there was a lot of stupid baserunning. Failed attempts at stealing third with no outs, failed attempts at stealing home. Just a lot of bad decisions. Players were running until they were thrown out or scored. Yes the scoring part is a positive but the out part is a negative.

Someone who can better play with databases could probably present the data better then I can, and show us exactly what kind of steals were being attempt in 1911. Whether they stole it or got caught in what kind of situation and then also show us a modern seasons stolen base numbers as well. I would love to do it but it would probably take me all day to get the 1911 part. I could probably do a modern season easily though.

This is good Ubiq. If many of those CSes are happening at 2nd with 2 outs, no big deal. If they are happening at 3rd with 0 outs, that is a big deal.

The value of a SB in a strict relationship to run scoring, is reductionist and misleading, IMO. The value of a SB/CS is extremely situation-specific. To look at all SBs and CSes and get a linear weight doesn't really mean much to me.

[Ubiquitous]

I'm looking at some of the data so far for 1911. It's a bit of a pain since I'm not good at databases (wheres Kent when you need him?) but so far with incomplete data it looks like the old players tried more higher risk steals then they do now and it didn't pay off so well.

For instance I found that a player stole third 72 times in 1911 all by himself. Meaning he wasn't part of a double steal he was the only one stealing a base. He stole the base 72 times but he was caught 88 times. I haven't figured out yet the outs situation.
Figures for stealing home are a little harder to figure because I think some of the caught stealing data is probably busted suicide squeezes or swing and misses and the runner breaking for home. Their were 5 solo steals of home (while a ton of double steals) and 59 caught stealing of home which lead me to the above conclusion that the data isn't strictly steals but busted plays as well.

There was 748 solo steals of 2nd base and 576 solo caught stealing of second.

In all there was 1029 events when people stole a base (more then one steal happened in some of these events there was almost 1700 steals) and 723 events in which somebody got caught stealing. Though keep in mind that while the stolen base data or at least the total is complete the caught stealing data is incomplete. AS of right now a huge chunk of PbP data is missing and it is likely that hundreds upon hundreds of caught steals are not recorded still. For instance Cincin had the second most steals but over half their games do not have CS data so they only have 61 cs, the fewest in the league. Which is doubtful, I wouldn't be surprised if there is 300 to 400 missing CS.

If anybody wants I have some of the base situation and event numbers for stolen bases and caught stealing on an excel file for 1911. If you want more you'll have to download the retrosheet files and do it yourself.

[Sultan_1895-1948]

Also I'll say it again the players back then were attempting different kinds of steals. Meaning it wasn't the same ratio of steals today. 85% steal of second, 14% steal of third, and 1% of home (just made those up). They were trying to steal third and home more often.



We should also keep in mind that the raw CS% isn't accurate (in the traditional sense), because many were delayed steals, walkoff steals, getting into pickles on purpose with first and third or getting picked off. Runners were just more daring, and there was a reason for it. Every run meant more, which meant it was worth the risk.

I see what you're saying about CS meaning just the same, but it's risk/reward. Two hits (sometimes three with outfielders playing so shallow) vs. one hit (bigger lead and secondary off second, although in rare cases, two hits might still be required).

Chris, have you chimed in on the hitting aspect of this discussion? Many times before I'm sure. Feel free to cut and paste ;) Unless of course, you're now in the GHR camp; then new documentation is required.

[538280]

We should also keep in mind that the raw CS% isn't accurate (in the traditional sense), because many were delayed steals, walkoff steals, getting into pickles on purpose with first and third or getting picked off. Runners were just more daring, and there was a reason for it. Every run meant more, which meant it was worth the risk.


I did Cobb's relative SB% for years we have data a while ago, and even that context of more people getting caught they're anything but impressive for the supposed greatest baserunner of all time.

[Ubiquitous]

Definitely in that era action was good. It was more then just exciting basball but it lead to runs. But the punishment for failure cost the same, and that really is my point. The value of an out didn't diminish, now it's possible that other events happened more often when running but again that doesn't change the value of an out. You can run around all you want cause as many errors as you can, steal as many bases as you want but if at the end of it you are thown out at home or third then it all doesn't matter. When a runner fails he loses the ability to have all those pluses happen for him. Now what I am seeing while looking at the stolen base data is that a great many runners were taking chances they should not have been taking. It seems a lot of guys had the philosophy of I'm going to run and keep trying for the extra base until you get me out. Which according to the data did happen quite a bit. Which leads me to think that part of the reason why run scoring was so low was because players were making lousy decisions on the basepaths. It would be interesting to check out the 1922 NL data that retrosheet has and see where the stolen bases were coming from. For instance in 1922 just 11 years later errors declined by 450. A 21% drop in just 11 years. Now is it possible that the decline in errors is because players stopped running around the basepaths or was the better fielding the cause of less running around the basepaths?

But anyway back to the 1911 data. Just take a look at some of the games, 4 caught stealings in a game is not an unusual occurence. I don't care what the environment removing 4 baserunners is going to hurt.

[Ubiquitous]

Here are some of the things we know.

In 1911 if you tried to steal a base chances are it was going to be a 50/50 shot on average. So right off the bat we already know that it wasn't easier to steal a base back then. It was harder. Why? Because runners were running when they shouldn't. Now then we would then need to know what happened in the half of the successful steals. How often did it result in an error and an extra base? I have some of that data, but I have to go pickup a pizza, I'll look at it later. But right from the get go the steal is starting off in the hole because it has a gigantic amount of outs already on its ledger. So the extra base bonus must be somewhat high or else it just isn't going to mean a whole lot in terms of value. If an error occurred on 20% of the steals and their was 1600 steals then we got 1600 outs plus 1600 bases plus we'll say 350 extra bases from errors. So we have about 2000 bases with 1600 outs. Is that really better then 700 bases for 260 outs (approximate modern steal value)? Is it adding runs or is it costing runs at that point?

[Ubiquitous]

For what its worth, I believe that Cobb had a greater ratio of runs/(hits+walks) than any of the players of his era (in his prime years). If his steals minus caught stealing were not a scoring benefit to his team that would not have been so.

This reminds me of something. How many ROE and FC did Ty Cobb have? I'm willing to bet he has a huge amount of missing times on bases to his stat-line. He might verl well have the most missing in the game.

[Sultan_1895-1948]

For instance in 1922 just 11 years later errors declined by 450. A 21% drop in just 11 years. Now is it possible that the decline in errors is because players stopped running around the basepaths or was the better fielding the cause of less running around the basepaths?


We need to keep in mind the changes that occurred from 1911 to 1922. Important things happened. The rules that created a harder ball which was easier to see for the entire game. Players began to copy Ruth's swing, and switched the thicker barreled, thinner handled bats. More hard hit balls were being put in play, and fielders realized that their gloves weren't cutting it anymore. So they went with bigger gloves that had more padding. Haven't seen any numbers on it, but I betcha with the harder hit balls, the double play numbers went up. Also, outfielders started playing deeper, creating more space for hits to drop in. Some field dimensions were altered to encourage homers, which in turn encouraged more fans to keep coming out.

This had to be the first instances where runners realized the risk/reward wasn't what it used to be :noidea

Maybe with the new rule changes, the catchers were then able to throw a more true ball down to second as well.