Probably not a coincidence that his best players are the ones who never played for him. That said, he was a true MICHIGAN MAN who would have had an EXCELLENT APR.
Anatomy of a Recruiter: Chalmers "Bump" Elliot
After my preliminary Diary, "Anatomy of a Recruiting Class: Bennie Oosterbaan's Last Class", I decided that I was going to take a different route in writing these. I am bunching teams together by coach, in order to give an idea of how each coach operated. I smoothed out my method of class analysis, as well. I am only considering freshmen, not walk-ons that start in their Sophomore or later seasons. As such, I actually wind up skipping a couple of early All-Americans.
The standard calculation I use for the capability of a recruiting class is the ratio of man-games started over total possible games played. Prior to 1965, there were 11 man-games played per game played, and after 1965 there were 22 man-games played per game played. Higher the ratio, better the class. In totaling how good a class was, I am using the formula:
M-G Ratio + .025 * All-Americans + .01 * Drafted Players + .1 * Heisman Players + .0001 * 3-year Players
From what I've come up with, a ratio of over .5 is considered to be a great class.
Chalmers "Bump" Elliot coached from 1959 to 1968, so his recruiting classes spanned the 1960-1969 Freshman classes. In his time as head coach, he only made one bowl game, in 1964.
1960: 60 Freshmen
1961: 24 Sophomores, 9 starts in 9 games.
1962: 23 Juniors, 37 starts in 9 games.
1963: 16 Seniors, 16 starts in 9 games.
1964: 4 Seniors, 10 starts in 10 games.
10 players made 72 starts, for a man-game ratio of .2136
15 players played for all 3 years of their eligibility.
Tom Keating, DT, Buffalo Bills, 1964
Class Score: .2251
1961: 61 Freshmen
1962: 25 Sophomores, 21 starts in 9 games.
1963: 21 Juniors, 48 starts in 9 games.
1964: 16 Seniors, 53 starts in 10 games.
1965: 0 Seniors
12 players made 122 starts, for a man-game ratio of .3961
14 players played for all 3 years of their eligibility
Robert Timberlake, QB, 1964
Robert Timberlake, K, New York Giants, 1965
Arnold Simkus, DE, Cleveland Browns, 1965
John Henderson, WR, Philadelphia Eagles, 1965
Class Score: .4525
Note: 1965 was the first year that offense and defense were separated, doubling the number of man-games played. There also were 10 regular season games from 1965 onward.
1962: 50 Freshmen
1963: 34 Sophomores, 25 starts in 9 games.
1964: 29 Juniors, 29 starts in 10 games.
1965: 27 Seniors, 46 offensive and 50 defensive starts in 10 games.
1966: 5 Seniors, 14 offensive and 23 defensive starts in 10 games.
16 players made 187 starts, for a man-game ratio of .3535
25 players played for all 3 years of their eligibility
William Yearby, T, 1964&1965
Jack Clancy, E, 1966
Jack Clancy, WR, Miami Dolphins, 1966
Thomas Mack, G, Los Angeles Rams, 1966
Steve Smith, OT, San Francisco 49ers, 1966
William Yearby, DE, New York Jets, 1966
Class Score: .4710
1963: 48 Freshmen
1964: 28 Sophomores, 18 starts in 10 games.
1965: 27 Juniors, 47 offensive and 45 defensive starts in 10 games.
1966: 25 Seniors, 54 offensive and 33 defensive starts in 10 games.
1967: 2 Seniors
17 players made 197 starts, for a man-game ratio of .2814
24 players played for all 3 years of their eligibility
Richard Volk, DHB, 1966
Mike Bass, DB, Green Bay Packers, 1967
Jim Detwiler, , Baltimore Colts, 1967
Frank Nunley, LB, San Francisco 49ers, 1967
John Rowser, DB, Green Bay Packers, 1967
Rick Volk, DB, Baltimore Colts, 1967
Carl Ward, DB, Cleveland Browns, 1967
Class Score: .3688
1964: 53 Freshmen
1965: 28 Sophomores, 18 offensive and 15 defensive starts in 10 games.
1966: 24 Juniors, 26 offensive and 28 defensive starts in 10 games.
1967: 21 Seniors, 39 offensive and 52 defensive starts in 10 games.
11 players made 189 starts, for a man-game ratio of .2625
20 players played for all 3 years of their eligibility
David Porter, , Cleveland Browns, 1968
Ray Philips, , New Orleans Saints, 1968
Rocky Rosema, LB, St. Louis Cardinals, 1968
Class Score: .2945
1965: 55 Freshmen
1966: 31 Sophomores, 15 offensive and 16 defensive starts in 10 games.
1967: 24 Juniors, 35 offensive and 23 defensive starts in 10 games.
1968: 22 Seniors, 41 offensive and 33 defensive starts in 10 games.
1969: 1 Senior, 9 offensive starts in 11 games.
12 players made 172 starts, for a man-game ratio of .2522
21 players played for all 3 years of their eligibility
Ronald Johnson, HB, 1968
Ronald Johnson, RB, Cleveland Browns, 1968
Tom Stincic, LB, Dallas Cowboys, 1968
George Hoey, DB, Detroit Lions, 1968
Class Score: .3093
1966: 50 Freshmen
1967: 30 Sophomores, 23 offensive and 25 defensive starts in 10 games.
1968: 33 Juniors, 40 offensive and 34 defensive starts in 10 games.
1969: 26 Seniors, 43 offensive and 44 defensive starts in 11 games.
1970: 3 Seniors, 2 offensive and 6 defensive starts in 10 games.
14 players made 217 starts, for a man-game ratio of .2924
22 players played for all 3 years of their eligibility
Thomas Curtis, S, 1969
James Mandich, E, 1969
Brian Healy, , Minnesota Vikings, 1970
Thomas Curtis, DB, Baltimore Colts, 1970
Garvie Craw, , Boston Patriots, 1970
Barry Pierson, , St. Louis Cardinals, 1970
Cecil Pryor, , Green Bay Packers, 1970
James Mandich, TE, Miami Dolphins, 1970
Class Score: .4046
1967: 36 Freshmen
1968: 23 Sophomores, 13 offensive and 20 defensive starts in 10 games.
1969: 17 Juniors, 39 offensive and 32 defensive starts in 11 games.
1970: 16 Seniors, 32 offensive and 34 defensive starts in 10 games.
1971: 1 Senior
11 players made 170 starts, for a man-game ratio of .2408
16 players played for all 3 years of their eligibility
Dan Dierdorf, T, 1970
Marty Huff, LB, 1970
Jack Harpring, , New York Jets, 1971
Jim Betts, , New York Jets, 1971
Don Moorhead, , New Orleans Saints, 1971
Marty Huff, LB, San Francisco 49ers, 1971
Pete Newell, , Detroit Lions, 1971
Dan Dierdorf, T, St. Louis Cardinals, 1971
Class Score: .3524
1968: 52 Freshmen
1969: 28 Sophomores, 24 offensive and 34 defensive starts in 11 games.
1970: 26 Juniors, 58 offensive and 56 defensive starts in 10 games.
1971: 21 Seniors, 81 offensive and 86 defensive starts in 12 games.
1972: 3 Seniors, 22 offensive starts in 11 games.
17 players made 361 starts, for a man-game ratio of .4826
21 players played for all 3 years of their eligibility
Thomas Darden, DB, 1971
Reggie McKenzie, OG, 1971
William Taylor, HB, 1971
Mike Taylor, LB, 1971
Paul Seymour, OT, 1972
Paul Seymour, TE, Buffalo Bills, 1973
Fred Grambau, , Kansas City Chiefs, 1973
Thomas Darden, DB, Cleveland Browns, 1972
Mike Taylor, LB, New York Jets, 1972
Reggie McKenzie, G, Buffalo Bills, 1972
Glen Doughty WR, Baltimore Colts, 1972
Tom Beckman, DE, St. Louis Cardinals, 1972
Mike Keller, LB, Dallas Cowboys, 1972
William Taylor, , Atlanta Falcons, 1972
Mike Oldham, , Washington Redskins, 1972
Guy Murdock, C, Houston Oilers, 1972
John Seyferth, , New York Giants, 1972
Class Score: .7297
1969: 36 Freshmen
1970: 23 Sophomores, 8 offensive and 4 defensive starts in 10 games.
1971: 19 Juniors, 36 offensive and 35 defensive starts in 12 games.
1972: 16 Seniors, 44 offensive and 47 defensive starts in 11 games.
1973: 3 Seniors, 10 offensive starts in 11 games.
15 players made 184 starts, for a man-game ratio of .2323
15 players played for all 3 years of their eligibility
Randy Logan, DB, 1972
James Coode, , Atlanta Falcons, 1974
Larry Cipa, QB, New Orleans Saints, 1974
Randy Logan, DB, Philadelphia Eagles, 1973
Bo Rather, WR, Miami Dolphins, 1973
Clinton Spearman, , Los Angeles Rams, 1973
Bill Hart, , Chicago Bears, 1973
Class Score: .3188
In order, the scoring of Bump Elliot’s classes is as follows:
Bump had 193 of 501 Freshman players use their full eligibility, 16 All-Americans and, 49 Drafted Players. His best recruiting class saw 5 (!!) All-Americans (1968), though they were awarded under Bo.
Tune in next time for an analysis of Bo’s classes, and their placement in the rankings.
It's worth remembering that the Big Ten still had their rule that the only bowl they could play in was the Rose Bowl. In retrospect, Elliott wasn't the greatest coach, maybe not even a very good one, but he did win that one Rose Bowl, 35-7. As a player, he was one of the "Mad Magicians," who went undefeated and beat USC 49-0 in the Rose Bowl.
It would be really nice to see UM beat USC 49-0 in my lifetime.
I think that using this analysis might finally put some numbers behind the "Lloyd could recruit but he couldn't develop his players" meme.
who were recruited late in his tenure, you can see why Bo was successful right off the bat. I wonder how Lloyd's last couple of classes compare and how that impacted RRod's first season (and perhaps the coming one as well).
John Henderson (Eagles and Bills 1965)
Mel ANthony (Browns 1965)
Chuck Kines (Bears 1966)
Paul Staroba (Browns 1971)
Bill Laskey was not drafted, but had a long NFL career, making the Pro Bowl one year. Bill Keating was also undrafted but played in the NFL.
In the early 60's Michigan ran separate offensive and defensive platoons (with occasional guys playing some on both sides of the ball). It wasn't until 1965, though that the starts were assigned separately. So, in 1964, for example the number of starts listed seems to depend on whether the offense or defense played first (with the exception of qb, where Timberlake is listed as starting all 10 games). Ward, Detwiler, Anthony, Mack, and Patchen only played offense and Yearby, Volk, Rindfuss, Laskey, Cecchini, and Nunley only played defense.
I excluded all players that were not part of a freshman class. Hence those draftees being missing. Thanks for the information for pre-1965, though. I'll keep that in mind.
What makes you designate them as "not a part of a freshman class?" All but Staroba appear as freshmen in the Bentley Library database. Not sure why Staroba doesn't appear, but the fact that he was starting as a sophomore makes you figure he was involved as a freshman.
I must have missed a few in my draft lookup on the NFL site then. Some people walked on after Freshman year. I excluded those. I'll look into those names later today to see why I missed them. If they didn't show up in my search on the NFL Draft History page, they aren't here. I'll do that tonight and make adjustments as necessary.
Edit: I added Henderson, had missed him, but the rest either do not show on the NFL history site or were walk-ons after freshman year. I'm only included drafted people that show up on the NFL site.
Mel Anthonhy doesn't show up on the official NFL site. It seems it is incomplete for late rounds in that era. He was taken by the Browns in the 16th round in the 1965 draft (actually held on Nov. 29, 1964).
Here is a link to a contemporaneous newspaper page which gives the entire NFL draft for that year and shows Anthony to the Browns. http://probaseballarchive.com/Viewer.aspx?img=17591891_clean&firstvisit=...
someone over at GBMW analyzes those numbers. i'm no good with numbers and junk.
Good research! I like this! Yay! Exclamation points!
I think you are overweighting starts. If a coach brings in 3 crummy classes in a row, some of those guys have to start. If he brings in 3 great classes in a row, no more can start than in the previous situation.
Oosterbaan recruited 4 guys that played in 6 pro bowls - Barr, Kramer, Morrow and O'Donnell. Elliott inherited none of them.
Elliott recruited 7 guys that played in 26 pro bowls - Laskey, Darden, Dierdorff, Ron Johnson, Tom Keating, Mack and Volk. Bo inherited 2 - Darden and Dierdorff.
Bo recruited 10 guys that played in 23 pro bowls - Bostic, Dave Brown, Carter, Jumbo Elliott, Haji-Sheikh, Harbaugh, Hicks, Hoard, Desmond Howard, and Kenn. Moeller inherited 1.
It's all relative. If the kid is the best on the team at that time, he will probably start the most. You can't deny the greatness of the 1968 class. I think adding draftees and all Americans to the math made it more accurate than if I had stuck with just starts.
in 1969 Michigan beat OSU. Bo gave the game ball to Bump.
(somewhere in a Bo book. Probably in Lasting Lessons)
is that it does not include a metric/variable for on the field performance... which was the typical gripe with LC... he had the players, they were drafted, played in the NFL.... but he did not use them on the field, especially in his last 6 years.....
but like any other metric driven analysis, your analysis provides good directional information and provides good perspective.
I can't really think of any good on the field metrics that are available... but I do think that starts is a good measurement. Someone wouldn't make 12 starts if they weren't the best on the team at that position at that time.
really blossomed under a new and presumably better coach...
My father used to tell me that Bump Elliot detested recruiting and felt that a player would come to Michigan on his own accord, or not...it was up to the player to decide and the coaches shouldnt have to recruit him.
Super Nintendo Chalmers?
"Me fail English? That's unpossible"
I understand you're trying to apply a universal strategy to a very diverse set of recruiting classes, but I'd like to see some sort of justification to the weighting you chose. I'm not necessarily arguing with it, but you don't provide any reasoning to justify it either.
I chose the weighting for the following reason:
Starts generally defines how good a player is relative to the rest of the team. If one class is full of excellent players, they will have more starts than those classes surrounding them. I can't think of a better way to do things than what I've come up with, really, except to add the ratio. I'm trying to think of a good way to ratio in the record of the team in a year to how many starts were made... but I do feel like my assessment works at least fairly well.
I'm not saying you're wrong, but there are a couple of ways I'd change the computation (I'd be interested in obtaining your dataset.)
In summary, I'd eliminate the arbitrary weighting. I'd acknowledge that the Heisman is heavily biased towards offensive players. I'd also acknowledge that the college game is sufficiently different from the pro game, and that the pros recruit players that fit their system.
I like your idea of rewarding starts, but I'd measure a different metric; effective wins (ie, season wins * ratio of starts to available starts.) I'd sum of each class's effective wins over their four (or five) seasons and determine an overall class score that could be used to compare classes.
In actuality, we should probably normalize to the number of games played (total wins * ratio of starts / games played) to account for the fact that the modern football season is longer than it used to be. There are a lot of interesting ideas to play with provided you have the data.