Diamondbacks 5, Giants 3 - Jeff the Giant Killer

by Jim McLennan on Sep 11, 2007 11:59 AM EDT in Game Reports Comment 43 comments

Record: 82-63. Change on last season: +14. Pace: 92-70
Playoff odds: 89.7%. Playoff Magic Number: 13

Quote of the day: " Eighty or ninety percent of me is not as good as 100 percent of Justin Upton or Carlos Quentin. I couldn't be selfish. But when I took batting practice, I felt fine hitting." -- Jeff Salazar

Kirk Gibson would be proud. Hobbling off the bench, Jeff Salazar smacked a three-run homer with two outs in the ninth, and limped his way around the bases. That extended Arizona's winning streak to six games - now four in a row, coming from behind - and guarantees us our first winning season since 2003. It was only Salazar's second ever four-bagger: curiously, the other one also came against the Giants, in September, and as a pinch-hitter, back when he was with the Rockies.

It was an improbable win, and an ironic mirror-image of the situation on the other side of town. In the NFL stadium, Arizona had the lead against San Franciso until almost the last possible moment, before coughing up a big, fat hairball. Basically, business as usual for both franchises: the D-backs find another way to win, the Cardinals snatch defeat from the jaws of victory once more. But who is that got the headline spot in the Republic this morning? A clue: not the team which is leading their division.

Justifiable bitterness aside, this is probably the best game I've never seen a single pitch of. I was working on...something I can't talk about (my 'Project Cloverfield', if you like!), and so was unable to see a pitch or even listen to the game. But for about 8 2/3 innings, it looked like the streak was going to end. Sure, we'd come back from three down, and were within one run, but I didn't bother going through to watch the game, even after the tying run, in the shape of Drew, reached second to lead off the ninth. I was content: hey, a five-game streak, that ain't bad.

That we needed to come back, was largely thanks to Benji Molina, who took revenge for the foul tips we whacked off his brother's helmet on Friday, by doing some whacking of his own. Two more homers off Hernandez 2.0, but as usual, they were solo shots. 31 home-runs this year, but only eleven came with men on base, and only three with more than one man aboard. Quality start for Livan: six innings, three runs, though as ever, with eight hits, three walks and the help of three double-plays. Business as usual there, too.

We didn't get on the board until the fifth, not helped by some gaffes on the base-paths. And by the time we did, the Giants had reached an 85.8% win probability, following Molina's second homer. Emilio Bonifacio, in his first career start, got his first career hit and his first career RBI, with a two-out single, which scored Upton from second. He repeated the medicine in the seventh, only with Montero on second, and also made a brilliant play to rob Roberts of a hit in the ninth. That was crucial, as any base-runner would have brought Bonds to the plate as the tying run in the ninth. Best avoid that, I think. Don't be surprised if Bonifacio gets more starts, probably in place of Callaspo. He just looks hungrier and happier to be there than certain back-up infielders I could mention.

Credit to the bullpen too, who did what they've been doing all streak long: keeping the team in games, and giving them the chance to fight back. Famine fanned the site in the seventh, Slaten and Peguero wobbled but posted a zero in the eighth, and Pestileñce brought the hammer in the ninth, retiring the side in order for his third career save. Whatever was ailing him in August has clearly been fixed, because here's his line in September:
Peña: 6.2 IP, 1 H, 1 BB, 5 K, 0 ER, .045 OBA
Yes, you read that right: twenty-two at-bats this month, one hit, a single.

Second-largest Gameday Thread ever, passing the 350 for the Padres on 8/27 and trailing only the 428 for the same opponents on August 30th. After Sunday's turnout, I'm relieved everyone hasn't abandoned me. We might have made 400 if carahan hadn't lost his way to the thread. :-) Welcome, none the less. Also present were: snakecharmer, suitsmetoATnT, mikeb (welcome!), LucaMaz3, DbacksSkins, hotclaws, seton hall snake pit, singaporedbacksfan, Diamondhacks, johngordonma, VIII, soco, DBACKS KICK ARSE, Peachy, peeklay, TwinnerA and FormerGiantFan. See? We got the win for you!

Gameday Graph

[Click graph to enlarge, in new window]
Master of his domain: Jeff the Giant Killer, +74.1%
God-emperor of suck: Tony Clark, -13.5%
Honorary "Well done!": Stephen Drew, +17.0%

0 recs | Comment 43 comments

Story-email Email Printer Print

Comments

Nice to see...

...that Sally gets to be Master of his Domain, with a staggering +74.1%.

Oh, and we're at 400 posts, as of this morning. :-)

Come on, Sandy Baby, loosen up. You're too tight.

by DbacksSkins on Sep 11, 2007 12:03 PM EDT reply actions 0 recs

snort

74%!!!!!

Have we ever seen a number like that before?

And just a hair off of 90% playoff odds. Wow, that's a beautiful thing. One good series against the Dodgers will make that number very strong!

by johngordonma on Sep 11, 2007 12:50 PM EDT up reply actions 0 recs

Not to mention

finishing off LA as a threat.

by peachy rex on Sep 11, 2007 12:56 PM EDT up reply actions 0 recs

I asked Jim a question like this awhile back.

I remember for sure that Drew hit something like 78% earlier in the year. (I think it was his homer that made Hoff blow the save)

I think somebody hit over 100% too, didn't they?

Come on, Sandy Baby, loosen up. You're too tight.

by DbacksSkins on Sep 11, 2007 1:00 PM EDT up reply actions 0 recs

Second-biggest score of the season

Stephen Drew got an +80.3% for his performance against the Padres on April 25th. That, and Salazar's shot last night, are the only two ninth-inning homers which took us from being behind to ahead: however, Drew's came with the tying run on first rather than third, so our win probability was lower [only 10.2%], and it was also a walk-off, so the win probability after it was the full 100% - the Giants still had a chance, albeit slim, of coming back last night. Indeed, because Drew had a negative score from his first three at-bats, his homer, by itself, was worth a full 89.8%. Link

by Jim McLennan on Sep 11, 2007 2:51 PM EDT up reply actions 0 recs

power rankings

we're #6 over at SI which is exactly where I'd put us:
http://sportsillustrated.cnn.com/2007/writers/john_donovan/09/10/power.rankings/index.html

But what happened to Dayn Perry? After mercilessly promoting our D'Backs before the season (I think he had us around #8 or so pre-season, which was about 8 spots higher than anyone else), he now refuses to give us props. Did an AZ girl break his heart?

http://msn.foxsports.com/mlb/powerRankings

We're behind both Detroit (HUH? You mean the same Detroit that is virtually guaranteed it won't make the postseason) and the Padres (you mean the Padres we just took two of three from and now have a 3.5 game lead on?).

by johngordonma on Sep 11, 2007 12:58 PM EDT reply actions 0 recs

Dayn Perry

I have been watching the foxsports power rankings all year, and it seems that Perry refuses to rank the D-Backs above the Padres...unfortunately there isn't even anywhere to comment on his inability to rank teams beyond his favorites, or an email address for him. I think SI has it right.

by AnybodyAnytime on Sep 11, 2007 1:49 PM EDT up reply actions 0 recs

Dayn Perry

Is run differential's bitch. He refused to give the team any credit in 2005, because they out-performed the expected value by about ten games. So he's probably sulking that the team is once again kicking run differential's ass, all the way from Arizona to Bill James' back patio.

by Jim McLennan on Sep 11, 2007 1:58 PM EDT up reply actions 0 recs

Dayn Perry

espn is showin no love either. they dropped them to eleven. i still dont see how they can keep using run diff and expected W/L to drop them even though they move up in the actual standings

by Adam1 on Sep 11, 2007 2:12 PM EDT up reply actions 0 recs

::Sigh::

As I just wore myself out railing against Dayn Perry, might I refer you to any of my patented tirades against that cesspool they call ESPN?

Come on, Sandy Baby, loosen up. You're too tight.

by DbacksSkins on Sep 11, 2007 2:20 PM EDT up reply actions 0 recs

Run Diff

I think run diff is fairly reliable. But give me a break, Perry. It's the middle of SEPTEMBER now... at this point win %, H2H, and current streaks have to be the major determining factors for rankings, NOT run diff.

by johngordonma on Sep 11, 2007 2:54 PM EDT up reply actions 0 recs

Using a Magic 8 Ball

I'm not a fan of run differential. i think its ok stat, but with holes. like say RBIs. but... Expected W/L is a joke. the only reason people believe in it is because its figured with an equation. People base how good they think a team is on this, but come on. at the end of the year, its rarely ever dead on. in 2006 only 2 teams matched their E W/L, 2 in 2005, and 4 in 2004. its just a guess. even MLB includes it in their standings now. whats next, will they use Dayn Perry's rankings to determine the postseason lineup?? who cares what a team should have won. the dbacks get no coverage because of run diff and E W/L, even though theyre a good, exciting team.

by Adam1 on Sep 11, 2007 3:13 PM EDT up reply actions 0 recs

LOL

Dayn Perry, while using his edjumicated method to determine this week's Power Rankings:

"Should the Dbacks be ranked even below the Padres?"
"All signs point to yes!"

Come on, Sandy Baby, loosen up. You're too tight.

by DbacksSkins on Sep 11, 2007 3:29 PM EDT up reply actions 0 recs

Run differential

Is interesting, but not much more. There is a good correlation to W/L record, but we already have a stat with an even better correlation. It's called "W/L record". :-) Otherwise, run differential's main flaw is that it assumes all runs have equal value, whether allowed by Brandon Webb in the ninth inning of a scoreless game, or Augie Ojeda in the ninth inning of a blowout. That is such an obvious fallacy to me, that it renders any conclusions drawn as a result of run differential, immediately highly suspect.

by Jim McLennan on Sep 11, 2007 4:09 PM EDT up reply actions 0 recs

Highly suspect ?

As the Dbacks pull farther away from their pythag, I'm certainly open to the idea that something other than luck is driving at least part of that discrepancy. I dont recall Bill James suggesting otherwise. But I disagree with your reasoning and conclusion that Run Diff is fundamentally "highly suspect".

The core statistical assumption isnt so much that individual runs have the same value, it's that over time, in a big enough sample, the game contexts under which those runs are scored tend to even out (migrate to a mean). The overall data clearly bear this out. Perfectly, as in zero variance for thirty teams at year end? Of course not.

But as of today, 26 teams are within 4 games of their pythag(plus or minus), 28 teams are within 5 games, and 29 teams are within 6 games. Over 140+ games, that's a damn good correlation. Well, except for....you know. That 30th team - the one whose GM finished a minus four last year. They're not seven up. Or eight games over pythag. Or nine games. Or ten games. Or eleven games. They're a dozen games over! Not twice the league variance. Twice that of the team with the second highest variance. Forget about playing at or near their pythag. The Diamondbacks could still have the highest positive pythag variance in baseball (at +7 for example), and be practically out of the wildcard.

The debate about what makes up the +12 is fun food for thought - large variances may even shed some new light on traditional sabermetric thinking. But before you crown Josh Byrnes with the genius tiara, remember that the pre-Josh 05 team (10 or 11 gms over pythag, I forget), had all sorts of bullpen issues and couldnt manufacture a run to save their lives, with entirely different personnel. These disparate anomalies are not easily explained by scientific roster makeup, or repeatable managerial acumen. Their occasional presence hardly discredits the fundamentals of Run Diff, any more than winning the lottery discredits the theory that winning the lottery is very unusual:-)

by Diamondhacks on Sep 11, 2007 6:31 PM EDT up reply actions 0 recs

What good is run differential?

All it does is provide a reasonably-good estimate of a statistic that we already know exactly - W/L. Teams that score more runs and concede fewer, will tend to win more games than those that don't. Well, duh. Pardon me if they're not quite handing out the Nobel Prize for that one. Using it to predict wins and losses is like listening to traffic to decide when to cross the road. There are simply better methods available - like looking at the actual W/L record!

If run differential was any good at predicting future records, it might be useful. But those are far more dependent on the runs scored and conceded in future, not what's gone before. If you can come up with some other use for run differential, beyond giving us an estimate of a figure we already know, I'd be interested to hear it. But until then, it seems no more than a statistical curiosity.

by Jim McLennan on Sep 11, 2007 8:20 PM EDT up reply actions 0 recs

usefulness

I think it is useful at the beginning of the season to determine who is "for real." 30, 40, 50 games into the season W/L % can be rather misleading and luck could have played a significant role into any given team's success or failure. Run diff (as well as strength of schedule) helps sort out the wheat from the chaff.

However, of all the "new" numbers (OPS, WARP, Win Shares, etc), Run Diff is definitely the least compelling, and once you get this far into the season you really have to throw it out. Sure, the D'Backs probably aren't as good as their record, but anybody who has watched this team knows that it is not by sheer, dumb luck that we've outperformed our Run Diff by +12. We are, without a question a far superior team than the Giants (who are basically even with us Run-Diff-wise). We are, without a question, a slightly better team than the Dodgers and Rockies (who have outperformed us Run-Diff-wise). And I would claim that we are a tad bit better than the Padres (who have also creamed our Run Diff).

By the way, in terms of future success, I do think Run Diff has some value (not a lot, but some). I would suggest that somewhere between the two numbers (win % and run diff) gives you a read on the future success of a team.

While AZ is an example of a team that was able (at least until now) maintain the ridiculous Run Diff rate, the Mariners are an example of a team who were way overperforming Run Diff and came crashing back down.

by johngordonma on Sep 11, 2007 8:41 PM EDT up reply actions 0 recs

Agree with paragraphs

1, 3 and 4, so let me challenge you on #2, just for fun :-)

We are, without a question a far superior team than the Giants (who are basically even with us Run-Diff-wise).

Really? I think we're somewhat better, but I'm curious how much better you really think we are. How many games? We know the actual W/Ls and run diffs, so let's step back from those & look for a different statistical way to evaluate the teams.

ERA+ and OPS+?

113 AZ 88
109 SF 90

According to this model, these are very comparable teams, in terms of both offense and defense. Is there anything, other than our explicit W/L, that makes you think we're the "far superior" team?

by Diamondhacks on Sep 12, 2007 12:32 AM EDT up reply actions 0 recs

starting pitching and bullpen

The San Fran offense is better than our offense. No question about that. However, that is somewhat compensated for by our superior baserunning/speed and superior defense.

The ERA+, however, as you know, doesn't tell the true story. First, SF's numbers are going to look better b/c of their home ball park and ours are going to look worse. Second, our rotation is top-heavy, with two very good starters and our fifth starters have been AWFUL. SF has a move even rotation, without much difference between 1 and 5. I would contend that any team would rather have a very strong 1-2 than a decent 1-5 (especially in terms of post season success).

Finally, our bullpen is far, far superior to SF's, even though, like our starters, the difference will be muted in the stats. We've had 2-3 bullpen arms this year have been AWFUL (headed by Nippert) and the rest have been superior. SF has a whole bunch of average arms.

I would indeed contend that our team is MUCH better than San Fran's.

by johngordonma on Sep 13, 2007 11:13 AM EDT up reply actions 0 recs

Maybe separately,

In pieces, our team is pretty even with San Fran's, but we're just better at putting those pieces together, in the right place at the right time.

Also, Bruce Bochy sucks.

I would point out that the cleanup guys (those lousy bullpen guys, headed by Nippert, EdGon, Piggy, etc.) are much worse when we're already behind. In the few situations where they pitch and we have a lead, they've mostly done splendidly. Mostly.

Also, John, I believe ERA+ takes into account ballpark pitcher biases, doesn't it??

Come on, Sandy Baby, loosen up. You're too tight.

by DbacksSkins on Sep 13, 2007 3:06 PM EDT up reply actions 0 recs

Jekyll & Hyde Pen - Luck or Design?

In pieces, our team is pretty even with San Fran's, but we're just better at putting those pieces together, in the right place at the right time.

That's really what I was trying to convey to John. Thanks.

I believe the Jekyll and Hyde bullpen does have something to do with our amazing pythag. It makes theoretical sense to me.

Counterpoint. Most if not all teams "try" to rotate their relievers in this fashion, more or less. Your top guys pitch when you're winning, your "long" guy comes in when you're behind,etc. To wit, the Dbacks advantage here doesnt seem to be strategic (ie scientific, reproducible) in nature. It's an advantage resulting from the fact that, this year, they "happen" to have crappy long men and four or five guys just lights out. I dont mean to suggest that's entirely coincidence, but the 180' opposite position - that this amazing, broad-based bullpen performance is the sole result of planning and philosophy - strikes me as perilously close to azphan's 3.6 +2 rubric, where no one on the staff is "allowed" to have an ERA over 3.6.

Do we really wanna go there?

by Diamondhacks on Sep 13, 2007 3:25 PM EDT up reply actions 0 recs

Well...

Our offense's general inability to come back after falling behind is a factor as well. We're usually not going to outscore anyone, so if we fall too far behind, the game is lost. Usually. We're hitting a little better these days, though.

I'm a little upset that this series screwed up our run differential a bit, (Going 2-1 and actually outscoring the competition by 6) but I think it highlights a point. Both Jim and Hacks have made good points about run differential: whether or not it's worth anything, since it only reflects the past, as does W/L; whether it accurately quantifies overall team performance; whether or not it can be used going forward. However, there's another side to it. The theory goes that, in a large enough sample size, (162 games) the ratio of run differential should approach win %. I think we can all agree that generally, this is true. We can also all agree that (duh!) W/L is what actually counts in the standings.

The pythag/run diff/whatever you want to call it theory is usually interpreted as "when a team over or underperforms its pythag, eventually its record will catch up to that." But I think, in our case, we might be seeing the flipside. Our win % has been consistently very good this year, even though we've been outscored. I think now, in the homestretch of the season, the team is showing that it's legitimately good, (even without O-Dawg) and actually our run differential is now evening out more because of that. Two nights ago, we had a 1 run loss; we won the other 2 games by a combined 7 runs. I honestly wouldn't be surprised if we finished the year having only been outscored by a run or two. Granted, that'd still be quite an accomplishment, and we'd still be outperforming our pythag, but not to the degree that we've seen most of the year.

The team is playing more conventional ball now. Maybe it's partly because the expanded rosters are taking away some of those quirks that made us such mathematical overachievers this season. Maybe it's because our A bullpen guys (Lyon, Peña, etc.) looked like they were running on fumes in August. Maybe it's because guys like Nippert, EdGon and Piggy have pitched much better. Maybe it's because we're hitting more. Most likely, it's all of the above.

Come on, Sandy Baby, loosen up. You're too tight.

by DbacksSkins on Sep 13, 2007 4:37 PM EDT up reply actions 0 recs

Know what else is useful?

I hear that if you calculate the combined win % of every pitcher on a team, you get the whole team's win %!!

Come on, Sandy Baby, loosen up. You're too tight.

by DbacksSkins on Sep 11, 2007 9:17 PM EDT up reply actions 0 recs

W/L

Ahahahaha. genius. now if only Dayn Perry had an email address....

by Adam1 on Sep 11, 2007 9:42 PM EDT up reply actions 0 recs

Oh, I'm sure he does.

It's just unlisted. Try dperry@foxsports.com or something? ;-)

Come on, Sandy Baby, loosen up. You're too tight.

by DbacksSkins on Sep 11, 2007 9:53 PM EDT up reply actions 0 recs

I think you kinda answered

your own question here:

future records...are far more dependent on the runs scored and conceded in future, not what's gone before.

That's true. A corollary is that future records are also more dependent on previous RS/RA than on previous W/L record. Seattle's a good example of both those true statements. So are the Yankees, whose early RS/RA was a much stronger predictor of future W/L than previous W/L, but not quite as strong a driver as second half RS/RA. Over the course of the season, this regression to the pythag mean is happening all over the leagues.

I indicated 26 teams were within 4 games of pythag after 140 games? Well, 26 teams were also within 4 games of pythag after only seventy games. That means there's more pythag volatility early(gap as % of total games played), but pythag generally approaches actual as the season progresses. The Dbacks, for whatever reason(s), obviously defy this convention in a big way, expanding their league high pythag gap all season.

As far as Nobel Prizes, the only real historical duh regarding RS/RA was on a per game basis. The team that scored the most runs on a given day won the game. Over the course of a season, that relationship was less obvious than you might think before Run Diff came along. Pretty smart baseball people would engage in results oriented analysis, pretending to explain why its team won games, when they were more accurately describing how they won. Hustle, bullpen, St Penelope or whatever.

By definition, Run Diff will obviously never surpass actual W/L in the limited way you described it - that was never its intention. It's purpose was to improve upon the results oriented conjecture that flew for years as gospel as to why specific teams win and others dont, by introducing a more scientific, verifiable (not perfect) way to look at the issue.

by Diamondhacks on Sep 11, 2007 11:51 PM EDT up reply actions 0 recs

Predicting the past

[ I indicated 26 teams were within 4 games of pythag after 140 games? Well, 26 teams were also within 4 games of pythag after only seventy games. ]

And all thirty teams were within zero games of their actual W/L record. Again, you're ignoring my question: what does run differential give you, except a poor approximation to W/L record, a stat we already have a perfect grasp on? I'm afraid, "to improve upon the results oriented conjecture that flew for years as gospel as to why specific teams win and others dont, by introducing a more scientific, verifiable (not perfect) way to look at the issue," doesn't actually cut it. How, exactly, is something which can be wrong by 12 games, and has no predictive value (because it's based entirely on past events) an improvement? After 70 games this year, here are two teams' records:

Team A: Run diff +51, Record 39-32
Team B: Run diff +65, Record 35-34

With similar run differentials, you'd expect them to have similar records going forward, no? Hasn't happened. Team A (Oakland) have gone 31-43 since, while Team B (NYY) are 46-28. Run differential tells us only about the past, which is as useful as choosing the division winners in October. No-one will be impressed with your abilities there.

The Yankees aren't winning now because they underperformed Pythagorean expectations in April and May. They are playing radically better now, even by the measure of run differential. April, their RD was +6. May, +18. June, +28. July, +78. Guess which month they had the best record? Again: well, duh. Outscore your opposition by almost three runs per game on average, and you will do quite well. Really, I'm still waiting for run differential to tell me something that isn't painfully obvious. Until then, it's basically a toy, and nothing more.

by Jim McLennan on Sep 12, 2007 2:14 AM EDT up reply actions 0 recs

Let's start

with a point of agreement. Current Run Diff is a better indicator of Current W/L than is Past Run Diff. No argument there. We also agree that the Yankees scored a lot of runs and the A's didnt since June, and that's reflected in their respective 2nd half records. But look at those teams again, back on June 20th. The A's had the better record - if W/L is such a good predictor of it's own future, why did Oakland fade so badly and the Yankees surge? June Run Diff was a better predictor of future W/L than was either team's W/L to date. Not as sharp an indicator as future Run Diff, to be sure, but better than past W/L. Again, I think you're answering part of your own question :-)

You seem to see Run Diff as this kind of no brainer tautology that is so obviously true that it doesnt have any value. Let's look at the relationship between Wins and Run Diff over time. Think of a game as a snapshot of Run Differential. The team with the most runs wins. Always. It's an absolute relationship. Now think of a series, or a week, as the snapshot. Well, Run Diff points to actual record, but it sure isnt an absolute correlation - teams win series all the time getting outscored. It's a positive but relatively volatile correlation. Snapshot of a month? More games (bigger sample), that volatility lessens and it becomes harder to "fool" run diff, winning the entire month with a negative differential. Snapshot a season, or five consecutive seasons, and "fooling" run diff becomes harder still (except if you're BoMel ;-)

Run Diff isnt saying that the Yankees offensive explosion wasnt significant. What it's saying is that, on June 20th, they were 5 games below Expected W/L,and that trend wasnt going to continue, regardless of how well or poorly they played the rest of the season. IOW, that 5 game negative gap wasnt going to become 10 games, then 15 games and so on. (FWIW, Oakland and NYY have both "stabilized" around 2-4 games below "expectations".)

It's kind of like an invisible hand, that says you're too far out of whack. It's not predetermined - no one's saying the Dbacks are going to crash 12 games under pythag in the final 18 games to come out "even". Not gonna happen. Their Sept. record will likely be most closely tied to their Sept RS/RA. Like any other team. But Adam Smith's hand is also out there, for everyone, and what it's telling the Diamondbacks is, "You can certainly keep winning, but not like you have been. Not forever."

by Diamondhacks on Sep 12, 2007 4:22 PM EDT up reply actions 0 recs

Luck be a lady tonight...

[ What it's saying is that, on June 20th, they were 5 games below Expected W/L,and that trend wasnt going to continue ]

Actually, no. Run Differential record has no memory. If you over-perform one month, you've still got just as much chance of overperforming the next month. Believing, "it's kind of like an invisible hand, that says you're too far out of whack," is like betting the house on black at roulette, because red came up the last five spins.

Certainly, over 162 games, there'll be a strong correlation between runs scored and runs conceded and wins. But the D-backs have over-performed their Pythagorean expectations every month this season. And pretty much did so two years ago, as well. I find it basically impossible to believe this is mere "luck" at work, and if there are other significant factors at play, then the value of run differential, whatever that might be, is significantly eroded.

by Jim McLennan on Sep 12, 2007 5:03 PM EDT up reply actions 0 recs

Gap Composition

If you over-perform one month, you've still got just as much chance of overperforming the next month.

That's right. 50% the first month, 50% the second month, 50% the third month. That means the odds of exceeding (or shorting) pythag three months in a row is 12.5%. The Diamondbacks are currently hovering in the 1-2% range.

I find it basically impossible to believe this is mere "luck" at work...

Me neither. I think when a team is this far off the charts, it's necessarily due to a confluence of factors - disparate things all pointing in the same direction. If it was just one or two simple variables, like a good bullpen and/or good pinch hitting, I suspect these large gaps would happen alot more often and more predictably. I think one of the factors here is "luck", and to deny that luck is at least a component of such a large gap strains credulity, IMO.

Melvin's two very high positive gaps in three years are, indeed, interesting. He's the only common denominator I can think of between those two teams. Different personnel, different strengths and weaknesses. Is he part of the reason for their great record in 1 run games both years? Could be, but Bob also fell short of pythag his other three years at the helm, so I'm somewhat hesitant to bestow instant praise on his game management philospophy when they're winning, while conceding his game decisions this year have worked out exceedingly well.

FWIW, my opinion is that the Dbacks are a better team than their run diff and that roughly half to two thirds of the pythag gap is probably luck that cant be scientifically reproduced. No proof, just a hunch based on watching team's pythags flip back n forth for a while now. If Josh can string together two or three or four of these, there'll certainly be reason to talk :- )

by Diamondhacks on Sep 12, 2007 6:11 PM EDT up reply actions 0 recs

great conversation

very smart contributions from both of you. I think I find myself somewhere in the middle. I do think that if you take any two teams with the same record (let's say us and the Yankees or Mets, who are the closest to us), Pythag is a helpful way to differentiate between the teams and predict future success. There's no question in my mind that we're a notch below those teams.

Now, there are two major problems with Pythag in predicting future success. The first is that Pythag only tells you about the team that has been on the field. It obviously doesn't tell you what will happen now that Hudson is out for the year or now that Hughes is back in the Yankees rotation or that Martinez is back in the Mets rotation.

The second problem is that, as Jim and others have pointed out: all runs are not equal. The fact that Nippert et al give up LOADS of runs in blow outs while our other relievers are incredibly stingy in tight contests does mean that those runs are less important and should count less in our pythag.

Finally, Pythag doesn't take into account certain debatable intangibles such as how "clutch" a player is. Now, I think that the ability for a player to be "clutch" is much exaggerated by the old guard, but I suspect that there are a handful whose focus and ability to have success increases in certain crucial situations. It doesn't seem to me that there is anyone who fits that bill on the D'Backs, but I could be wrong.

by johngordonma on Sep 13, 2007 11:28 AM EDT up reply actions 0 recs

I dont fully understand

what these power rankings are supposed to measure, exactly. a) Best team right now? b) Best team all year? c) Most likely division winner?

Regarding the relative merits of the Padres and Diamondbacks, the Dbacks are the better team right now - the hotter team - especially since Chris White Young's injury and dubious return.

In terms of the better team all year, I think quite a strong case can be made that the Padres have been the better team and just havent caught as many breaks.

Projecting a division winner (ie projected final record)is largely derived from what one thinks about a & b. Personally, I agree with John that it's getting late in the game and I would rank the Dbacks ahead right now - but they're still just two up in the loss column with 18 to go, and if you truly think San Diego is a better team (based on b, run diff,etc) I dont think it's unprofessional or outrageous to rank them a spot ahead.

by Diamondhacks on Sep 11, 2007 3:56 PM EDT up reply actions 0 recs

Biting and probably fair observation.....

but I also give him credit for sticking to his methodological guns. He talked up the Dbacks early, based on x,y, and z. And after x and y didnt happen and they won anyway, he didnt abandon his rubric, just to tell everyone, "I told ya so".

I respect that.

by Diamondhacks on Sep 11, 2007 3:30 PM EDT up reply actions 0 recs

Are you

happy at all the kids are doing so well? Any excitement at all?

by unnamedDBacksfan on Sep 11, 2007 3:35 PM EDT up reply actions 0 recs

This can be summed up in 5 words.

Dayn...Perry...is...an...idiot.

Who is this guy, some pothead from San Diego?? Firstly, look at all the teams he's ranked first this year: Boston, Anaheim, (Sorry... Los Angeles Angels of Anaheim) Cleveland, Yankees, Tigers. Yankees?? Tigers?? At least John Donovan's been consistent: The BoSox have been on top all year, (as they should be) except for a short period when they were replaced by the Angels, and a very early stint in first place by the Mets.

Oh yeah... and Donovan has the brains to rank the Dbacks above the Padres, after the latter just got their asses kicked by both the former AND the Rockies. Oh, and don't let your biases show or anything, Perry. I know the Dodgers are a game ahead, but does anyone REALLY think LA has the better team right now than Colorado?? Donovan's got the balls to rank the Rockies 10th, and I've got no problem with that. Might I remind Mr. Perry that the Dodgers just lost series' to the Giants and Padres, (sandwiched in between an oh-so-impressive showing by taking 3 of 4 from the Cubs... oooh, the Cubs!!) and the Rockies just beat the Padres AND Giants in subsequent series'?

Oh, and for just a final word on his ranking of the Padres above us, I quote his blurb about SD: "...and Brett Tomko has been added to the fold." He actually considers that a GOOD thing?? At GLB they were basically EMBARRASSED that they picked up Brett Tomko, he of the 2-11 record and 5.67 ERA.

Moron. It's no wonder he's working for FoxSports.

Come on, Sandy Baby, loosen up. You're too tight.

by DbacksSkins on Sep 11, 2007 2:14 PM EDT up reply actions 0 recs

Dont be shy now...

tell us how you really feel! ;-)

by unnamedDBacksfan on Sep 11, 2007 2:18 PM EDT up reply actions 0 recs

Actually

I really like Dayn Perry. I respect his journalism, his writing, his clearly paid attention to the game, and his opinions.

Come on, Sandy Baby, loosen up. You're too tight.

by DbacksSkins on Sep 11, 2007 2:23 PM EDT up reply actions 0 recs

JUST...EFFING...KIDDING

Dammit, now I'm so angry I can't read Arabic.

Come on, Sandy Baby, loosen up. You're too tight.

by DbacksSkins on Sep 11, 2007 2:24 PM EDT up reply actions 0 recs

Oh yeah

I forgot to even mention that he put the Mariners at 9th. 9th??? Those same Mariners who just got done playing themselves out of a postseason spot??

And look at his picture. What a fat, greasy turd.

Come on, Sandy Baby, loosen up. You're too tight.

by DbacksSkins on Sep 11, 2007 2:27 PM EDT up reply actions 0 recs

Giants fans at McCC were griping about this too...

'Why does Bochy continue to play first baseman Ryan Klesko, a free agent who doesn't figure in the club's future plans? The manager said he wanted to go with experienced hands against a crafty pitcher like Livan Hernandez. "We're trying to win games and Ryno has done a nice job for us," Bochy said.'

--San Jose Mercury News

Typical dumb Boch move. Can't see the forest for the trees. Of course, it wouldn't help the youngsters at all to get EXPOSED to a "crafty pitcher like Livan Hernandez" now, would it??

Come on, Sandy Baby, loosen up. You're too tight.

by DbacksSkins on Sep 11, 2007 6:14 PM EDT reply actions 0 recs