Eddie never met a walk he didn't like. :-)

what was his defensive reputation like when he was playing

what was his defensive reputation like when he was playing

He was certainly the king of 3B-putouts. I have no idea how that was viewed while he was playing but there has been some discussion about how to handle those today (I believe WS simply ignores them).

Part of his putout totals is due to his incredible durability. He played 838 straight games before tonsilitis finally got him.

That's certainly not the whole story with the putouts, though. Pitcher-handedness? Size of foul territory in WAS? Or was he just simply that good?

I would guess his POs consisted of an unusually large number of pop fouls. Wasn't there a lot of room in the Washington park? Or else considering the Senators pitching staffs in those days, a lot of screaming line drives. WS has him as a C- and that at least is good enough for me to conclude he was not Brooks Robinson.

Combined with a career OPS+ of 109 and only 4 years above 120, he's just not top 50 (probably not top 100) material. James has him #24, which is generous, considering he has him above a bunch of guys who get timelined to oblivion--John McGraw #26, Larry Gardner #29, Lave Cross #33, Harlond Clift, Denny Lyons, Billy Nash... I'd see him more like in the 30s somewhere among 3Bs, whereas we will be fortunate to elect 20 of 'em.

From the Win Shares Book: "Putouts by Third Basemen," pp. 228-230.

James concluded that PO by 3rd basemen were not a meaningful indicator of 3rd basemen's fielding ability. He looked at PO several different ways using a database of 1946-1960 teams.

He looked at determining expected PO vs. realized PO and found there was little consistency from year to year between players or teams. Also, there was little consistency from year to year among the third basemen with good defensive reps from the era (Kell, Boyer, Malzone, Carey, Keltner, etc.) in how they performed vs. expected putouts.
At other defensive positions, the comparison between expected PO or A and actual PO or A (adjusted for pitcher handedness and other things) is a repeatable skill. There was consistency with 3rd base assists however.

He did a second study using Gold Glove 3rd basemen from 1965 to 1990 and comparing their team 3rd base defensive numbers with a control team from each league each year and with the team from each year/league that had no regular at third. That is the team with the fewest games started by the player who played the most at 3rd.
James found the A and E totals were different, but that PO were the same between all three groups.

In summary, PO by 3rd basemen have no predictive value/do not appear to be a repeatable skill when comparing expected PO and actual PO. PO by 3rd basemen do not correlate to players with good or bad defensive reputations. And there is no team difference in 3rd base PO when one team has a GG'er there vs a team with no regular at the position. Therefore, he did not include it in his ratings for 3rd basemen.

"You can find a little bit of skill input here, if you look at it just right. But even then, the skill input is a third the size of the random variance, and a tenth the size of the variance in the category which can be atributed to other known biases. That's useless."

Those are James' opinions about 3rd base PO, at least I think that is what they are.

Then the fact that Yost got lots of them year after year is the fluke of the universe (next to humankind, anyway)? Still curious to hear any theories about that.

I wouldn't say they were a fluke.

In the section I mentioned above, James found that the major factors involving PO at 3rd were a pitching that didn't strike anyone out, was a flyball staff, played in a park with big foul territory, and playing for a bad team. Once those factors are accounted for, a player's actual PO performance on a year-to-year basis compared to expected PO is not a repeatable skill. One year a player is plus 15, the next minus 18, then plus 3, etc. This contrasts with, for example, expected double plays by team. That the Stengal Yankees were always over their expected totals is evidence that there is some repeatable skill involved involved and not just luck.

Now, could Yost's PO mean something? Sure. It is just that James found PO by 3rd basemen did not bear any identifiable relationship to fielding excellence - good/bad/middling.

Looking at some of the factors relating to Yost:

Was strikeout totals 1948-1956:
1948: 7th with 446. Lg avg. 538
1949: 7th with 451. Lg avg. 546
1950: 6th with 486. Lg avg. 570
1951: 7th with 475. Lg avg. 574
1952: 7th with 574. Lg avg. 644
1953: 8th with 515. Lg avg. 614
1954: 7th with 562. Lg avg. 641
1955: 6th with 607. Lg avg. 676
1956: 7th with 663. Lg avg. 727

Was Lg finish:
1948: 7th / 56-97
1949: 8th / 50-104
1950: 5th / 67-87 (25 games out of 4th)
1951: 7th / 62-92
1952: 5th / 78-76
1953: 5th / 76-76
1954: 6th / 66-88
1955: 8th / 53-101
1956: 7th / 59-95

re: foul territory at Griffith. I don't know for sure, but I think they were considerable.

re: GB/FB. I don't know where to find this.

But, three of the four factors for PO totals are present in Yost's case.

Hope this helps.

In the section I mentioned above, James found that the major factors involving PO at 3rd were a pitching that didn't strike anyone out, was a flyball staff, played in a park with big foul territory, and playing for a bad team.

how would playing for a bad team effect a third basemans putouts?

I don't know, but it was listed.

Can someone explain his leading in putouts in 1959 despite his team finishing 4th place and 2nd in strikeouts

More times 1st and 2nd occupied, I'd guess.

here's tiger stadium and Griffth Stadium

Can see how a bad pitching staff leads to 3B POs--runners on base, more thrown out at 3rd. And fly ball staff, a 3B is not gonna get a PO on a ground ball very often but a fly ball staff may mean more pop ups or more runners tagging up, or whatever.

James' factors are no Ks and fly balls, makes sense, but, again, just a bad pitching staff would seem to be important. Maybe he used "bad teams" as a surrogate for a bad pitching staff. Hard to understand that one otherwise.

But it's also hard to understand Yost apparently making POs "repeatable-y" if there's no repeatable skill. To me it has to come down to foul ground...?

Well, he did have 39 more PO than Harmon Killebrew did at 3rd that year. That year, only 3 players had more than 100 games played at third. Yost did average the most PO per game, but the Killebrew and Malzone had more assists, by 66 and 98, and Malzone just about doubled Yost's double play total, 40 to 21.

Yost leading in PO in 1959 is solely a function of health and ballparks. AL third basemen made 1279 PO in 1959. There were 1236 games including ties. There were 1.035 PO at 3rd each game. Yost had 168 in 146 games or 1.15 per game. That's good. But, Yost getting all those assists is mostly explained because he played the whole season at 3rd. The Tigers had 177 PO at third, but KC had 186 and Bal had 173. All three teams averaged the same PO per game, but Detroit was the only one with a regular 3rd baseman. It is the Roy Smalley Factor, but applied to PO rather than assists.
The other two teams with a regular 3b were Boston, which has NO foul territory, and Washington, where Harmon Killebrew played 150 games at third. Boston's regular 3b, Frank Malzone, had 98 more assists compared to Yost in only 8 more games, and only 34 fewer PO despite the smaller foul ground.
That Edd Yost led the league in PO at 3rd in 1959 means nothing more than he was healthy in the right ballpark.

To quote: Now, could Yost's PO mean something? Sure. It is just that James found PO by 3rd basemen did not bear any identifiable relationship to fielding excellence - good/bad/middling.
In other words, throughout 45 years of baseball, 1946-1990, James found no correlation between the number of PO by a 3rd baseman and the excellence of the fielder.

Could their be an exception every now and then? YES. It does not have to be perfect correlation EVERY year.

I would guess that Yost took "discretionary" PO. Pop flies that the catcher or SS could have taken, he did. Someone could compare the PO totals of the catchers/SS/3rd basemen of Yost's teams to the other teams in the AL during his career. But, first you would need to account for strikeouts and foul territory for each park each year. Then, you need to account for men on first because with a man on first the shortstop has to keep the double play in order.

What we need are accurate play-by-play reports of every AL game during his career with the location of the catch of every batted ball on the field like we have now. We don't have them.

A major tool James uses for his defensive measures is expected _____. Either double plays, assists, or PO. Using various numbers such as men on base, innings pitched by left- or right-handers, men on first, and park factors, to name a few, James will come up with an expected total. For example, if a team allows a lot of walks and singles, they would expect to turn more double plays than a team that didn't allow many baserunners. If both teams turned the same number of double plays, the first could be below expectation and the second could be above.
James found that players at most positions tended to carry over results from year to year. The results were not random. One team did not go from beating expected double plays significantly to missing the expectation the next year without there being a change in personnel. The results repeated themselves.

With third base PO, this did not occur. It is the only position on the field where James does not use PO. Players would go from exceeding expecations on 3rd PO to being below to exceeding. A player thought to be a good defender would not consistently exceed expectations on 3rd PO. On the other hand, at EVERY OTHER POSITION, good defenders would show themselves by exceeding expectations with assist/PO/double plays. (And not every year, just consistently.) THEY DID NOT DO THIS AT THIRD BASE REGARDING PUTOUTS.

A player's PO were much more impacted by ballpark, pitcher strikeouts, and playing behind a flyball or groundball staff, then playing for a bad team. Way below that, similar to the weight of a fly compared to the weight of a truck, to quote James, is the impact of defensive skill on 3B PO.

It would be interesting to know how much baserunners thrown out contributes to IF POs and expecially 3B POs.

Also I wonder about 2Bs and SS. It seems to me that their POs would also depend on 1) a bad pitching staff that put lots of runners on, and 2) a good DP partner. I mean, I wonder what percentage of 2B and SS POs are forces at 2B where the real work was done by the partner? I'd guess a ton. So the consistency of POs from year to year may really reflect the consistency of the DP partner?

The tail end of Yost's career is on retrosheet. His last league-leading season is here.

I checked some of the high-PO games and it does look like a lot of pop-outs. I only checked a couple of games, though.

Also, I'm not sure if retrosheet is a good source for the degree-of-difficulty of the outs. Often, its real vague like "made an out to third" but sometimes its specific enough to say "popped out to third" or "popped out to third in foul territory".

"Combined with a career OPS+ of 109 and only 4 years above 120, he's just not top 50 (probably not top 100) material."

I see what you are saying mark, but OPS+ is going to seriously underrate a guy like Yost - whose "O" to "S" ratio is probably as high as just about anyone's. He was a better player than OPS+ would lead one to believe.

Not saying James didn't overrate him (I'm not sure yet), but just cautioning against the use of OPS+ for a guy like him.

I would guess his POs consisted of an unusually large number of pop fouls. Wasn't there a lot of room in the Washington park? Or else considering the Senators pitching staffs in those days, a lot of screaming line drives.

a lot of screaming line drives would mean he was playing a lot closer to shorstop then third base.

About the putouts - WARP sure doesn't think much of it either. It gives him a staggering -162 FRAA, once it adjusts for all-time.

Was he really that atrocious? How does Win Shares view his fielding? A passing grade there might get Yost on the bottom of my ballot.

From Win Shares, p. 226:

"The Win Shares system would essentially agree with the Linear Weights analysis that there is little evidence that Yost was a quality defensive third baseman. Among third basemen with 10,000 career innings, he ranks as the third-worst ever, ahead of Richie Hebner and Pinky Higgins (although Dean Palmer, who is far worse, will cross the 10,000 inning mark early in the 2002 season).

Yost's letter grade rating is a C- (Palmer's a D-, Higgins a D, and Hebner a D+)

Also, IIRC, Yost's teams had an unusual number of lefty starters, including four in one season, creating more opportunities to field not only popups but also liners and grounders.

what was his defensive reputation like when he was playing
now that Yost is closing in on my pHoM I'll like to know if and how much WS underestimates his fielding based on contemperoy observers
since posting this despite searching I have yet to find any primeray source's on his fielding if anybody knows the answer or where I find the answer can they please post it

My only real memory of Yost was that he couldn't hit a lick, but as "The Walking Man" he walked a lot, and that when they played the SSB he flashed one of the chromiest domes in the Majors.

Oh, and BTW IIRC Griffith's Stadium's foul territory was not large at all, but rather quite small. I have no idea how to explain Yost's putout totals.

As is mentioned above, Yost was never out of the lineup. At the same time, most teams in the league did not have an everyday thirdbaseman. Looking at 1950-54, there were only 14 AL players with 125+ games at 3B: Yost 5, Rosen 4, Kell 2. Here's the year-by-year:

1950
157 Kell
155 Yost
154 Rosen
1951
154 Rosen
152 Yost
147 Kell
1952
157 Yost
147 Rosen
1953
154 Rosen
152 Yost
136 McDougald
1954
155 Yost
148 Boone
136 Finigan