Larry Walker’s Credentials Bear Repeating

On this year’s Hall of Fame ballot, four former players saw at least a 10-point increase in their voting share over the previous year. Vladimir Guerrero sailed into the Hall of Fame, Edgar Martinez solidified his status as a near-lock for next year, and Mike Mussina looks like a strong candidate for the 2020 class, if not the 2019 one.

Larry Walker, on the other hand, needs a lot of help. He received just 34.1% of the vote this year, leaving just two more cycles for him to reach the 75% threshold required for election. It’s not just that Walker needs some help to get elected: he wants it, too. And, most importantly, he deserves it.

Paul Swydan previously made a good case for Walker’s inclusion in the Hall, comparing him very favorably to Vladimir Guerrero. Here, though, I’d like to directly address a few points that still seem to cause confusion.

He Wasn’t Just Good Because of Coors Field

A lot of the arguments for Larry Walker’s inclusion in the Hall — including on sites like this one — are based on his very impressive 68.7 WAR. That figure ranks 66th all-time among position players and 39th since Jackie Robinson broke the color barrier. His WAR is sixth in that time among right fielders, just behind Reggie Jackson and ahead of every other right fielder you can think of except for a handful of all-time greats in Hank Aaron, Roberto Clemente, Al Kaline, and Frank Robinson

Walker doesn’t lack for impressive numbers by traditional measures. He has a lifetime .313 batting average, for instance, behind only Clemente, Vlad Guerrero, Tony Gwynn, Stan Musial, and Ted Williams among outfielders who’ve recorded at least 8,000 plate appearances over the last 70 years.

Walker also fares well by counting stats. He hit a lot of home runs, a ton of doubles, and stole over 200 bases. The list of players with more doubles, triples, homers, and stolen bases is a pretty small group, composed of just Aaron, Carlos Beltran, Barry Bonds, Andre Dawson, and Willie Mays. He won an MVP award in 1997, receiving more than three-quarters of the first-place votes from the writers. He also has seven Gold Gloves, five All-Star appearances, three Silver Sluggers. He earned at least one vote for MVP in eight different seasons.

There are those who might dismiss Walker’s accomplishments out of hand simply because of Denver’s thin mountain air. It’s certainly true that, with regard to the counting stats, some mental adjustment is necessary. As for estimates of his overall value, though, such considerations are irrelevant: WAR already penalizes Walker for whatever benefits he received from playing half his games at Coors Field.

From 1995 to 2004, a period that encompasses the latter half of Walker’s prime and nearly all his decline years, Walker played almost all his home games in Colorado. His overall slash line, home and road, during that time was .332/.425/.616. His wOBA — a metric that incorporates all his singles, doubles, triples, homers, walks, and outs — was .438 during that same period. Only Barry Bonds produced a higher figure among players with at least 4,000 plate appearances in that 10-year timeframe.

Even after adjusting for Coors, however, Walker was still excellent. Weighted runs created plus, or wRC+, is essentially the park-adjusted version of wOBA, and Walker’s 147 wRC+ ranks 10th within the aforementioned sample. Walker, in other words, was still one of the top hitters of his generation.

Or, here’s another way to think of it. Assume, for a moment, that Walker had produced the same numbers while playing at a totally neutral park. In that case, Walker’s line would translate into a 165 wRC+. Accounting both for hitting and baserunning, that translates to 440 runs above average. Walker’s overall career value in that case would have transcended 80 WAR. In reality, of course, that didn’t happen. Walker did play at Coors. So he recorded a 147 wRC+ instead and “only” 316 runs above average on offense. He produced just less than 70 WAR. Coors, in effect, “cost” Walker more than one WAR per season.

Of course it’s appropriate to consider Coors Field when discussing Walker and is necessary to downgrade his numbers somewhat. Don’t do it twice, though. Keep in mind that WAR has already lopped off 20% of Walker’s total value during his time in Colorado. And he was still great. Discounting Walker’s statistics even further is unnecessarily harsh.

He Was on a Hall of Fame Path Before He Arrived in Denver

Walker recorded the best season of his career during his time in Colorado and gained the most notoriety while with that club. That said, Walker was playing on a Hall of Fame level before he even joined the Rockies, while he was still a member of the Montreal Expos.

There are a couple ways to illustrate this while also demonstrating that Walker’s production in Colorado was just a logical extension of his time in Montreal.

Consider: from 1990 to -94 (age 23 through 27), Walker averaged more than four wins per season, with a very good age-27 season in 1994 cut short by the strike. If we take his 14.4 WAR from the three years before he got to Colorado, prorate the strike-shortened campaigned, and weight the years accordingly (50% for 1994, 33% for 1993, and 16.7% for 1992), we come up with an estimated talent level of 5.3 WAR for Walker’s 1995 season, his first campaign in Colorado.

That’s not a bad estimate, turns out: Walker posted a 4.6 WAR in the strike-shortened 1995 campaign. Prorated to a full season, that’s equivalent to… 5.3 WAR exactly. And, assuming a decent aging curve, a player who recorded 5.3 WAR in 1995 would be expected to produce roughly 45 wins over the 10 seasons from 1995 to 2004. That is almost exactly what he actually put up during his time in Colorado.

So, yes, alternate realities are impossible to predict. On the other hand, it seems that, even if Walker had gone to someplace that wasn’t Colorado, he could have been expected to perform as he actually did perform — which is to say, at a Hall of Fame level.

But that’s not the only way to arrive at that conclusion. Perhaps you prefer to answer the question using comps instead. With that in mind, I searched for outfielders from 1947 to -92 with 20-25 WAR from the age of 23 to the age of 27 who, like Walker, had at least 400 PA and a 120 wRC+ at 27. There were 14 outfielders who match those criteria. Of those 14, five are in the Hall of Fame. Of those 14, eight were corner outfielders and half of them (Jim Rice, Harmon Killebrew, Billy Williams, and Dave Winfield) are in the Hall of Fame. Another group (Roger Maris, Dave Parker, Reggie Smith, and Darryl Strawberry) were also impressive, if not Hall-worthy.

Of the players who played like Larry Walker when he was in Montreal, those who aged well made it to the Hall of Fame. Those who did not, fell short. Walker aged well and fits in nicely with the former group.

His Peak Compensates for His Lack of Longevity

WAR already does a pretty good job making the point above, but let’s take a look at the last 10 corner outfielders inducted into the Hall of Fame. Here they are, along with Walker.

Larry Walker and Hall of Fame Corner Outfielders
Name PA HR AVG OBP SLG wRC+ Off Def WAR
Rickey Henderson 13346 297 .279 .401 .419 132 650.9 -56.1 106.3
Carl Yastrzemski 13991 452 .285 .379 .462 130 463.3 -0.5 94.8
Reggie Jackson 11416 563 .262 .356 .490 139 499.7 -185.0 72.7
Tim Raines 10359 170 .294 .385 .425 125 408.2 -109.6 66.4
Tony Gwynn 10232 135 .338 .388 .459 132 401.5 -92.2 65.0
Willie Stargell 9026 475 .282 .360 .529 145 457.5 -185.0 62.9
Dave Winfield 12358 465 .283 .353 .475 128 406.5 -243.9 59.9
Andre Dawson 10768 438 .279 .323 .482 117 230.5 -11.8 59.5
Vladimir Guerrero 9059 449 .318 .379 .553 136 370.2 -115.0 54.3
Jim Rice 9058 382 .298 .352 .502 128 297.5 -106.3 50.8
AVERAGE 10961 383 .292 .368 .480 131 418.6 -110.5 69.3
Larry Walker 8030 383 .313 .400 .565 140 427.9 3.5 68.7

I’ll note, first of all, that comparing Walker’s numbers to the average line produced by this group of 10 might actually be unfair to Walker himself. Yastrzemski and Henderson rank second and fourth in plate appearances, respectively, since 1947. They very possibly skew the average upwards. In terms of median performance, Walker actually compares even more favorably than this.

Even so, Walker makes up the performance gap relative to this group. His wRC+ is worse only than Willie Stargell’s. In terms of runs above average, he is right there with the Hall of Famers. He doesn’t have the same playing time as the rest, but he still comes out the same in total WAR, and that’s because of Walker’s defense.

Look at that list above and ask how many of those players were good at defense. Yastrzemski was, and that’s what helped him to become an inner-circle Hall of Famer. Andre Dawson spent a bunch of time in center field to help his numbers. Guerrero had a good reputation but never won a Gold Glove and ended up spending nearly a quarter of his career at designated hitter.

Walker ends up at the same place value-wise because he was better on offense and defense than most of the guys above. That helps him compensate for the relatively shorter career, for retiring at 38 after an above-average season (135 wRC+) instead of hanging around a few more years without putting up much value. Edgar Martinez has a better batting line (147 wRC+) with around the same number of plate appearances as Walker and he is likely to be elected next year. Voting for a slightly worse hitter with vastly superior defensive skills hardly seems like a stretch.

*****

Larry Walker needs a 41-point jump in the next two years to make the Hall of Fame. It would be a rare feat, but not an unprecedented one. Around 50 players since World War II have been elected in their third ballot or later, and roughly 10% made the same jump Walker requires in just two years time. The average jump from two cycles prior (where Walker is now) to getting elected (where Walker hopes to be in two years) is 24 points, which would put Walker at around 58% in two years.

Since the end of World War II, eight players have gone from under 50% to the Hall in two years, four players have gone from under 40% to the Hall in the same timeframe, and five players have added more than 40 percentage points in just two election cycles. Below is collection of the biggest two-year jumps since 1946.

Big Vote Jumps in Hall of Fame History
Player Year Elected
to HOF
Voting % Two Years
Before Election
HOF Election % Change
Herb Pennock 1948 20.3 77.7 57.4
Hank Greenberg 1956 38.5 85.0 46.5
Joe Cronin 1956 33.7 78.8 45.1
Pie Traynor 1948 32.2 76.9 44.7
Luis Aparicio 1984 41.9 84.6 42.7
Harry Heilman 1952 51.8 86.8 35.0
Barry Larkin 2012 51.6 86.4 34.8
Tim Raines 2017 55.0 86.0 31.0
Eddie Matthews 1978 48.7 79.4 30.7
Jeff Bagwell 2017 55.7 86.2 30.5

Hank Greenberg, Joe Cronin, and Pie Traynor didn’t face the sort of circumstances in their own candidacy as Walker does in his, but there they are. None of the modern players here, meanwhile — Bagwell, Larkin, and Raines — benefited from quite the boost that Walker requires, but they came pretty close.

Even if Walker’s potential move up were unprecedented, it might make sense: the backlog of deserving candidates and the influence of the 10-vote limit on support for Walker are also unprecedented in their way. In many years past, if a player received 34% of the vote, it is because only 34% of the voters have felt that he deserves one. Because of the high number of potentially deserving candidates over the past decade, however, it would be rather presumptuous to assume that only 34% of the voters believe Walker is a Hall of Famer given. Half of the voters fill out a full ballot and are prevented by the Hall from voting for all candidates they feel are deserving.

A fair number of voters will certainly have to change their mind on Walker before he is enshrined. After a couple larger Hall of Fame classes, however, and the subsequent clearing of the backlog, he likely already has a decent number of voters who might be able to vote for him for the first time next season.

Larry Walker has done everything that is generally expected of a Hall of Fame player performance-wise. He has solid counting numbers, hardware, and great WAR numbers as well. He was already on his way to a Hall of Fame career before he went to Colorado. He had 11 more great years after that and went out near the top of his game without padding a few worthless seasons to the end of an already great career. Walker needs some big gains the next two years to get into the Hall of Fame, but he deserves it.

We hoped you liked reading Larry Walker’s Credentials Bear Repeating by Craig Edwards!

Please support FanGraphs by becoming a member. We publish thousands of articles a year, host multiple podcasts, and have an ever growing database of baseball stats.

FanGraphs does not have a paywall. With your membership, we can continue to offer the content you've come to rely on and add to our unique baseball coverage.

Support FanGraphs




Craig Edwards can be found on twitter @craigjedwards.

newest oldest most voted
stever20
Member
Member
stever20

Larry Walker Rockies Career
590 home games- 154 homers 42 hr pace
580 away games- 104 homers 29 hr pace

Sorry but that’s just a huge gap. The only reason why his counting numbers are even solid is Coors. He has 335 homers instead of 383 and he’s not even close.

southie
Member
southie

Le sigh

sadtrombone
Member
sadtrombone

I don’t know how much time you spent with those numbers, but if you’re just pulling the last four years of his career from the splits tool on Fangraphs (which is all that is up) he played that last year in St. Louis, where his wRC+ had a 60 point gap between home (164) and road (104) in wRC+. More specifically, he whacked 50% more homers at home, in a notorious pitchers park (9) than on the road (6) in the same number of at-bats. Would you like to blame that on Coors too?

This is besides the fact that a 29-home run pace in your last four seasons is generally considered pretty good, that home runs are not the only measure of worth, that home/road splits aren’t very accurate ways of capturing stadium effects…

stever20
Member
Member
stever20

Just looking at BR who has season splits- and using their OPS+ tool-
95 H 189 A 127
96 H 212 A 39
97 H 199 A 213
98 H 218 A 141
99 H 253 A 131
00 H 165 A 103
01 H 222 A 158
02 H 192 A 149
03 H 166 A 109
(the 96 number is not a typo for away- he had a away slash line of .142/.216/.307). The average taking away 96 and MVP year 97(best/worst)- 131. Last year had no 131 OPS+ guys- but had Hosmer, Rizzo, Arenado 132 Khris Davis 130. His home numbers same years- 201. (Trout was at 187)

I just see his numbers being way too Coors inflated. I’d be pretty sure voters have as well. Also, if you look at his numbers- he only hits 40 once and 30 only 4 times. Too short of a prime. I think it’s pretty easy to see why he isn’t in.

Expo27
Member
Member
Expo27

I actually went through and used the BR splits to manually calculate his road wRC+ during his Coors years a while back. I’m not at my home pc, so I can’t show my work, but it came out to 132 (.281/.385/.515), which is almost exactly the average for HOF corner outfielders in the table above. That assesses him no benefit for a possible Cooks hangover, or for the fact that almost every player hits better at home. No matter how you slice it, he produced at a hall of fame clip. Voters might think his numbers too Coors inflated, but by the standards they themselves have established they’re objectively wrong.

sadtrombone
Member
sadtrombone

I have some real questions about the park adjustment used here if adding the humidor affected a supposedly park-adjusted statistic (as you mention below), and how you’re weighting each of those seasons (that should be a weighted average). And you can’t really just throw out a season with 300 PAs and say it is equivalent to one in which he had 664 PAs. But let’s take that all that at face value.

Walker spent a little over half of his playing career in Colorado, about 4500 of his 8000 or so PAs. What you’ve calculated covers about 3500 of that. About half of that was at home, so about 1750 PAs. If you’d like to lop 1/3rd of the value (which is the difference between his home and road sOPS+ that you’ve calculated here) for those 1750 PAs, that takes the total amount of value down by 30 runs, which penalizes total value by about 7%, and which takes the total amount of WAR down from 68.7 to 63.9.

(I think this is not actually correct because of position adjustments, though I don’t have that one handy…and it’s very close anyway)

So before taking into account the Coors hangover, the fact that many players perform worse on the road, the fact that this isn’t a weighted mean, that this doesn’t have any effect on his MVP year in Colorado (since that was excluded from that analysis), that the Park factor obviously isn’t working properly since the Humidor has an effect on his home/road splits, and that Expo27’s calculations actually have Larry Walker’s penalty at more like 3.5% than 7%…if you take all of that into account then you wind up with a guy who is slightly below Tim Raines and Tony Gwynn instead of slightly above him.

The problem you’re making is that you’re noticing these dramatic differences (which themselves are likely overstated) that only affect a fraction of a fraction of a fraction (etc) of his value.

stever20
Member
Member
stever20

1 thing I’d say also looking at his splits from BR- the 2002 and 2003 seasons at Coors were the smallest gaps of his career. Oh, and I looked… The Rockies started using the Humidor in…. 2002 https://www.denverpost.com/2017/05/13/colorado-rockies-baseball-humidor-coors-field/

Llewdor
Member
Llewdor

How do Walker’s career road numbers compare to the average Hall of Famer’s road numbers?

Since most players play better at home, we need to do that correction for all players if we’re going to make a fair comparison. Road numbers only, what are the typical HoF credentials?

Handsome Wes
Member
Handsome Wes

Jaffe’s JAWS has Walker right at the average line for HOF Right Fielders.
https://www.baseball-reference.com/leaders/jaws_RF.shtml

Here are the away splits for Dave Winfield and Vlad, who are below the JAWS line:
.286 BA / .356 OBP / .485 SLG
.312 / .376 / .536

Walker’s #’s outside of Coors:
.282 BA / .373 OBP / .500 SLG

So he’s a comparable OF to those two.

alacy52
Member
alacy52

It’s not exactly what you’re asking, but Walker’s career road OPS (.865) is slightly better than Griffey’s career road OPS (.860).

francis_soyer
Member
francis_soyer

Other notable road OPS from approximately that era:

Vlad: .912
Kent: .857

And some worse defenders:

McGriff: .887
Edgar: .926

And a little more recently:

Berkman: .940

dcweber99
Member
dcweber99

And yet he still was outstanding in the outfield.

dl80
Member
dl80

I don’t know if we have reputable statistics to do him justice, but he comes out almost exactly at zero for his career on defense.

ballskwok
Member
Member
ballskwok

To be more accurate, he comes out at 3.5 runs *above average* on defense for his career, which is very good for a corner outfielder (that is after the positional adjustment). The list of corner OFs who have 8,000 or more PAs and above average defensive value is actually quite short (I count 10 between LF and RF, though I’m not sure the positional breakdown is accounted for in the leader boards (some players with time spent in CF, etc. appear to be included in the LF/RF leaders).

If you just look at all OF (including CF) with a minimum of 8,000 PAs, there are only 21 who have above average defense (Walker being the 21st on the list).

francis_soyer
Member
francis_soyer

Yeah, I was surprised to see Vlad at -115 for his career, so 3.5 is pretty legit.

RonnieDobbs
Member
RonnieDobbs

I would love a discussion of the quality of the data that Larry Walker’s defense is based on. Not saying he wasn’t a quality defender, but quantifying something without valid data is just deception.

jfree
Member
jfree

Actually Coors penalizes WAR on defense too. Coors has a huge outfield (probably the 2nd biggest at the time behind KC) – but it is adjusted like a hitters park (all small outfields).

He had a phenomenal arm. But at Coors that just doesn’t matter statistically. Maybe there is some data showing the batters/runners who didn’t stretch an extra base because he was in right field. But even if there is it would get drowned out by the high run scoring environment at Coors – which is a statistical outlier and normal distribution statistics (eg mean or standard deviation) SUCK at dealing with systemic outliers.

RonnieDobbs
Member
RonnieDobbs

Can someone summarize why this data is wrong? I see 75 down-votes as of this moment, so it is must be pulled from thin air (pun intended). I get that there are arguments for more complex analysis, but are the splits correct?

TKDC
Member
Member
TKDC

I think the point is that we have ways to look at Coors Field with nuance and correctly adjust for the advantage, and his response is basically “Derr, more home runs at Coors, derr…”

He basically just cuts off the 50 home run difference, which is a dubious thing to do considering plenty of guys do better at home than on the road, and then says his “counting numbers” wouldn’t have been close, which also presupposes that we should be setting counting number cut offs below which you are ineligible for the HOF.

RonnieDobbs
Member
RonnieDobbs

I was asking about the raw numbers. It sounds like those are correct. Whether you agree with his take or not (not hard to follow why you disagree), he points out a stat that people are going to hold against him.