Clay "Shad" Roe

[Vris]

In doing a quick historical replay using the Lahmen 5.2, I discovered a pitcher, Clay "Shad" Roe that imported with a "stuff" rating over 200! Yes I had it show ratings over 100. From 1927-1940 (I haven't done 1941 yet) he has gotten 300 strikeouts 13/14 years (1939 he got injured in June with only 253), 400 Ks 10 times, 500 Ks 6 times, and 600 Ks once! He is approaching 7000 for his career and has won 12 POTY awards in that time frame, including when he got hurt. He has a career ERA of 1.93 in almost 5000 career innings.

I find this insane and wonder, during a historical league, if anyone has approached anything like this. I don't know if it was something I did or what.

[MogulChamp]

Heh, can't say that I have. Sometimes when playing historical seasons players that hardly even played get pretty good stats and do a lot better than they ever did.

Here's his career stats according to my Baseball Encyclopedia:

Clay Roe 1923 Was 0-1 1.2ip 0h 6bb 2k

That's it. He never pitched again in the majors.

[Vris]

So it was my fault. I should've been following things more closely. I was just going and importing quick and checking things along the way. When I noticed he had 400 strikeouts in one year, I made sure it wasn't Nolan Ryan and thought it was interesting. Thanks for checking up on that.

[tcblcommish]

Originally Posted by Vris

So it was my fault. I should've been following things more closely. I was just going and importing quick and checking things along the way. When I noticed he had 400 strikeouts in one year, I made sure it wasn't Nolan Ryan and thought it was interesting. Thanks for checking up on that.

Nope not your fault at all. Statistical problems will arise when using the base Lahman's database. Because of the players like that only have very little playing time, they will always have bloated stats like that.

Try doing a search for Ankit's database in the mods forum and use it. you'll loike the results I think. I know I do. he does an amazing job.

[Vris]

I've used Ankit's before and was pleased, but this was really the first time I spent any real time on a historical league. I used Lahmen's 5.2 because I wanted to try out a historical quick sim and went with the "newest". I had never heard of Clay Roe and when I saw he was just dominating the AL I knew something was fishy.

[Comedian2004]

Since he only pitched in one game, if you had gone with the defaults, it would have created him lousy stats, right?

I always wonder about them one game players. That guy came in, did not allow a hit, but walked 6 batters and threw 3 wild pitches. Was he the owner's nephew or something?

[JDOldSchool]

Originally Posted by Comedian2004

Since he only pitched in one game, if you had gone with the defaults, it would have created him lousy stats, right?

I always wonder about them one game players. That guy came in, did not allow a hit, but walked 6 batters and threw 3 wild pitches. Was he the owner's nephew or something?

Nope, but his grandson might be Rick Ankiel.

[rudel.dietrich]

Only problem with the Ankit database is that a player must play the entire year to be on the roster for that year.
Bo Jackson does not show up until 1987 even though he had 90 some AB in 1986. Duke Snider does not show up with the Dodgers in 1947. Nor does he show up in 1948. In fact since he was a midseason callup he does not show up in the database at all that I can find. Those are just two examples. Im sure others are in there somewhere.

[rogmax11]

With a name like Clay Roe there was no way something wouldnt be fishy.

[M's rule]

Originally Posted by Fallschirmjager

Only problem with the Ankit database is that a player must play the entire year to be on the roster for that year.
Bo Jackson does not show up until 1987 even though he had 90 some AB in 1986. Duke Snider does not show up with the Dodgers in 1947. Nor does he show up in 1948. In fact since he was a midseason callup he does not show up in the database at all that I can find. Those are just two examples. Im sure others are in there somewhere.

From Ankit's readme file:

8. If a batter had at least 4,500 ABs in his career, his debut year is now the first season in which he had more than 238 ABs.

9. If a batter had at least 1,350 ABs in his career, his debut year is now the first season in which he had more than 130 ABs.

Point 8 covers why Snider doesn't import until 1949, even though he played in 93 games and had almost 250 AB in 1947 and 1948 combined.

Point 9 covers Jackson.

Point is, the database isn't flawed because of these omissions, because they're intentional (and in keeping with what the database attempts to accomplish). I guess it was inevitable that some players got seemingly misrepresented when those parameters were used--but my guess is that even if Ankit had decided to set those limits at different numbers, the same thing would happen to other players that now happens to Duke and Bo (and who knows how many more).

Aside from adding data to the csv files to create a 1947 and/or 1948 version of Duke Snider, or importing the data from Lahman when you're at that year (which has its potential for complications, since Ankit and Lahman are not compatible), I don't know of any way around this difficulty.

[rudel.dietrich]

Originally Posted by M's rule

From Ankit's readme file:

8. If a batter had at least 4,500 ABs in his career, his debut year is now the first season in which he had more than 238 ABs.

9. If a batter had at least 1,350 ABs in his career, his debut year is now the first season in which he had more than 130 ABs.

Point 8 covers why Snider doesn't import until 1949, even though he played in 93 games and had almost 250 AB in 1947 and 1948 combined.

Point 9 covers Jackson.

Point is, the database isn't flawed because of these omissions, because they're intentional (and in keeping with what the database attempts to accomplish). I guess it was inevitable that some players got seemingly misrepresented when those parameters were used--but my guess is that even if Ankit had decided to set those limits at different numbers, the same thing would happen to other players that now happens to Duke and Bo (and who knows how many more).

Aside from adding data to the csv files to create a 1947 and/or 1948 version of Duke Snider, or importing the data from Lahman when you're at that year (which has its potential for complications, since Ankit and Lahman are not compatible), I don't know of any way around this difficulty.

I don't think I ever said they were flaws.
It's just something I do not like about the database.

No perfect database exists.