The Odds are not so odd anymore.

[batmagadanleadoff]

Mets ESPN playoff odds are only 8% better than the Braves, even though they're two games up on the Braves with just nine games left to play.


Padres 87-66 (+2) (99.5%)
Mets 85-68 (--) (63.3%)
DBacks 85-68 (--) (82.0%)
Braves 83-70 (-2) (55.2%)

The DBacks are given an almost 20% better chance of making the playoffs than the Mets even though the two teams have identical records and the Mets hold the tiebreaker advantage over the DBacks.

Must be that looming head-to-head series with the Braves, right? (But the DBacks and Padres end their seasons playing each other). Or that the Mets finish the season with a tough schedule, at least on paper, playing three playoff contenders.

Padres 89-66 (+2) (99.9%)
DBacks 87-68 (+1) (93.5%)
Mets 86-69 (--) (69.1%)
Braves 84-71 (-2) (37.5%)

With the Braves unable to gain any ground on the Mets over their last two games, and the Mets reducing their magic number and the number of games left to play by two, ESPN now predicts the Mets to be almost twice as likely to make the playoffs as the Braves -- 69.1% to 37.5%.

[Lefty Specialist]

Those 3 games in Atlanta are basically the season. I'll look at the odds after that.

[batmagadanleadoff]

Those 3 games in Atlanta are basically the season. I'll look at the odds after that.

Yeah, but more so for the Braves than for the Mets.

[Lefty Specialist]

I've been scarred by too many late-season collapses to look at the odds until they're 3 ahead of the Braves with 2 to play.

[batmagadanleadoff]

Mets ESPN playoff odds are only 8% better than the Braves, even though they're two games up on the Braves with just nine games left to play.


Padres 87-66 (+2) (99.5%)
Mets 85-68 (--) (63.3%)
DBacks 85-68 (--) (82.0%)
Braves 83-70 (-2) (55.2%)

The DBacks are given an almost 20% better chance of making the playoffs than the Mets even though the two teams have identical records and the Mets hold the tiebreaker advantage over the DBacks.

Must be that looming head-to-head series with the Braves, right? (But the DBacks and Padres end their seasons playing each other). Or that the Mets finish the season with a tough schedule, at least on paper, playing three playoff contenders.

Padres 89-66 (+2) (99.9%)
DBacks 87-68 (+1) (93.5%)
Mets 86-69 (--) (69.1%)
Braves 84-71 (-2) (37.5%)

With the Braves unable to gain any ground on the Mets over their last two games, and the Mets reducing their magic number and the number of games left to play by two, ESPN now predicts the Mets to be almost twice as likely to make the playoffs as the Braves -- 69.1% to 37.5%.

The odds got odd again. With the Mets winning tonight, thus shaving their magic number against the Braves by one and passing the DBacks in the standings, their odds to make the playoffs shrunk a little -- while the Braves odds got better, by about 10%

Go figure.

Padres 90-66 (+3) (>99.9%)
Mets 87-69 (--) (67.2%)
DBacks 87-69 (--) (86.0%)
Braves 85-71 (-2) (46.8%)

[RealityChuck]

Because the "odds" are not mathematical odds. They are the results of a simulation. Garbage in, garbage out.

[batmagadanleadoff]

Because the "odds" are not mathematical odds. They are the results of a simulation. Garbage in, garbage out.

This is what I thought. But I figured that they must have simulated the rest of the season like a million times to arrive at their odds. I'm guessing that what's going on is that their model has the Braves as being a slightly better team than the Mets and winning the Mets/Braves series more often than the Mets in their simulations.

[Edgy MD]

It's ESPN. ESPN isn't so much a journalism outlet, but a brand. And now they've got a sports book affiliated with that brand, so I wouldn't make too much of odds from there, as it's in their interest to queer the method by which they arrive at those odds.

Not saying these odds are necessarily influenced by the gambly-gambly side of ESPN, but I don't trust 'em.

Apart from the day-to-day record, internal personnel things can be moving the needle — the return of Albies, the apparent setback of Senga, the Mets using Diaz for four innings over two days, the Braves scheduling Fried over Morton against the Mets — but that's just speculation.

[kcmets]

I've been scarred by too many late-season collapses to look at the odds until they're 3 ahead of the Braves with 2 to play.

Yeah, I mean not to be a Doubtin' Deborah Downergal but would
anyone be super surprised if the Braves swept this series?

[Benjamin Grimm]

Of course not. That's why we're heading into this series with this feeling of suspense.

[Edgy MD]

Of course, I'd totally blame kcmets if that's what goes down.

[Johnny Lunchbucket]

The pressure to win tonight will be insanely intense. It's a Really Big Game (R)

[Centerfield]

Nightmare scenario.

Lose the first two. Get rained out. End up tied after the Brewers series. Go back down to Atlanta and lose one more.

[batmagadanleadoff]

Mets and Braves ESPN odds to make the playoffs are now practically even after the Braves took the opener of their three game series. Mets (78.0%), Braves (69.2%). Probably because the Braves need just a split of the remaining two games to secure the tiebreaker and Sale likely to pitch one of those two games. Let's see what kind of havoc the Hurricane weather wreaks on the rest of the series.

[batmagadanleadoff]

Mets and Braves ESPN odds to make the playoffs are now practically even after the Braves took the opener of their three game series. Mets (78.0%), Braves (69.2%). Probably because the Braves need just a split of the remaining two games to secure the tiebreaker and Sale likely to pitch one of those two games. Let's see what kind of havoc the Hurricane weather wreaks on the rest of the series.

The good news, according to the ESPN odds anyways, is that the Mets, at 78%, are the likeliest of the three remaining NL contenders to make the playoffs. The DBacks are predicted to be the odd team out, what with both the Mets and the Braves holding their tiebreakers over Arizona.

[batmagadanleadoff]

ESPN odds now has the Braves as the odd team out. Their odds to make the playoffs dropped by about 10% even though both the Mets and Braves didn't play yesterday. The Mets odds stayed the same.

[rchurch314]

ESPN odds now has the Braves as the odd team out. Their odds to make the playoffs dropped by about 10% even though both the Mets and Braves didn't play yesterday. The Mets odds stayed the same.

Is there any public methodology to ESPN? What are they measuring?

Fangraphs has Mets at 77.1%, Diamondbacks at 61% and Atlanta at 61.9% Passes the eye test because the Mets have 1 less loss than both, Atlanta holds the tie-breaker against the Diamondbacks and Arizona does not have the benefit of games against their competition. The only reason Diamondbacks are not even lower is that even though it seems like they don't control their own destiny, both Mets and Atlanta cannot win out.

[Edgy MD]

Yeah, FanGraphs odds have always smelled the best to me.

[kcmets]

Postseason Probability
Morning of September 26th

DET 93.4%
KCR 85.0%
NYM 76.2%
ARI 73.4%
ATL 50.4%

-- baseball-reference.com

[Edgy MD]

I tried to put together a table to get all the numbers in one place, and it looks like ESPN (who I expressed doubt in) and FanGraphs (who I endorsed) are working with the same math.

Shows how much I know.

TEAM	ESPN	BB-REF	FANGRAPHS	AVERAGE
NYM	77.8%	76.2%	77.8%	77.3%
ARI	60.5%	73.4%	60.5%	64.8%
ATL	61.6%	50.4%	61.6%	57.9%

[rchurch314]

well, ESPN doesn't EXPLAIN where I can see, where this stuff comes from. Par for the course with them of course. I wonder if they're flat out stealing it. Even if they duplicated effort, randomness might have it slightly different.

[batmagadanleadoff]

ESPN odds now has the Braves as the odd team out. Their odds to make the playoffs dropped by about 10% even though both the Mets and Braves didn't play yesterday. The Mets odds stayed the same.

Is there any public methodology to ESPN? What are they measuring?

Not that I know of. I'm assuming that ESPN (and Fangraphs and BbRef) probably create digital Strat-o-Matic like cards for each of the players and then run a monte-carlo style simulation of the rest of the season, maybe like a million times, to arrive at their odds. Their models are probably sophisticated. Perhaps they factor in pitching matchups, recent injury trends and general recency trends, too.

[batmagadanleadoff]

Braves 87-71 (--) (85.6%)
Mets 87-71 (--) (57.3%)
DBacks 88-72 (--) (57.1%)

The Braves are heavy favorites to make the playoffs, among the three teams battling for the last two NL spots. Makes sense as they hold tiebreakers over both the Mets and the DBacks, and so, for all practical purposes, are a half a game ahead of both teams even though they are all tied in the standings.

The Mets and DBacks are essentially even, odd-wise, to make the playoffs -- the Mets only slightly ahead of the DBAcks even though they hold the tiebreaker there.

What I can't figure out at all, is how all three teams could be given a better than 50% of making the playoffs when this is now a game of musical chairs and when the music's over, one team is guaranteed to be left on the outside, looking in.

OE-- I guess it's possible. Because two of the three teams will qualify, the combined odds for all three teams should add up to 200%. But it doesn't seem right at first blush, even though it is.

[rchurch314]

ESPN odds now has the Braves as the odd team out. Their odds to make the playoffs dropped by about 10% even though both the Mets and Braves didn't play yesterday. The Mets odds stayed the same.

Is there any public methodology to ESPN? What are they measuring?

Not that I know of. I'm assuming that ESPN (and Fangraphs and BbRef) probably create digital Strat-o-Matic like cards for each of the players and then run a monte-carlo style simulation of the rest of the season, maybe like a million times, to arrive at their odds. Their models are probably sophisticated. Perhaps they factor in pitching matchups, recent injury trends and general recency trends, too.

I know how the other two do it. They're real stats pages. They do run it thousands of times. BBRef uses more of recent record, Fangraphs factors in projected results off the depth charts/matchups/strength of schedule.

ESPN is probably stealing it, unaccredited.

[batmagadanleadoff]

Who knows. It's more of a toy, as far as I'm concerned. I assume everybody else thinks the same.

ESPN's odds also has the Brewers as favorites to win tonight's (Saturday) game 57.5% to 42.5%. Also ESPN's opening betting money line has the Brewers at (-125) -- that is, if you're betting the Brewers, you must bet $125.00 to win $100.00.