Lineup Discussions

[Bellhorn04]

No, he really can't. Beating one's Pythagorean W-L is not a sustainable skill. It's another one of those fluky random things.

 

I love numbers, but if Pythagorean record can be out of whack for as long a time frame as 4 years, it really doesn't mean very much IMO. Sure it works as a general correlation-common sense tells us it must-that's about it.

[southpaw777]

Toronto 2 games out

Baltimore 3 games out

MFY only 4 out now

 

Im not too worried about the MFY, even if they make the playoffs. A lot of kids have fairly good succes when they first come up because nobody really knows them or has a book on them. I think thats part of the reason they are surging Like the sox did when Farrell left and Lovullo played the kids. If they get in, they wont go far at al IMHO.

The good news is the rivalry should start to heat up a bit between the two teams again come 2017.

[Kimmi]

I love numbers, but if Pythagorean record can be out of whack for as long a time frame as 4 years, it really doesn't mean very much IMO. Sure it works as a general correlation-common sense tells us it must-that's about it.

 

Technically, it hasn't been out of whack for 4 years.

 

No "predictor" in baseball is going to be extremely accurate. We all know that. Pythagorean W-L does give a better indicator of remaining season performance than actual record, however. At least until late August/September.

 

My point here is that the Yankees are 'lucky' to have as many wins as they have. The ideas that 'a good team will find a way to win the close games' or 'a good team can regularly beat its Pythagorean W-L' or 'a good manager can regularly lead his team to beat its Pythagorean W-L' are all false notions.

 

Beating Pythagoras is not a repeatable skill. It is luck or randomness.

[a700hitter]

Farrell stinks, and I hope the Sox keep winning.

I always liked the guy.

[Bellhorn04]

Technically, it hasn't been out of whack for 4 years.

 

No "predictor" in baseball is going to be extremely accurate. We all know that. Pythagorean W-L does give a better indicator of remaining season performance than actual record, however. At least until late August/September.

 

My point here is that the Yankees are 'lucky' to have as many wins as they have. The ideas that 'a good team will find a way to win the close games' or 'a good team can regularly beat its Pythagorean W-L' or 'a good manager can regularly lead his team to beat its Pythagorean W-L' are all false notions.

 

Beating Pythagoras is not a repeatable skill. It is luck or randomness.

 

OK, it's luck or randomness. But apparently it's a form of luck or randomness that can be sustained for a long time.

 

If I'm flipping a coin it might come up heads the first 5 times in a row, but if I keep flipping it will correct to 50% very quickly.

 

A good hitter might be able to hit .400 for 50 games, but he will regress to .300 by the end of the season.

 

But a team who beats their Pythagorean record by 5 games might very well do it again the following season.

 

The correction factor seems weak to me. That is my real point.

[Kimmi]

OK, it's luck or randomness. But apparently it's a form of luck or randomness that can be sustained for a long time.

 

If I'm flipping a coin it might come up heads the first 5 times in a row, but if I keep flipping it will correct to 50% very quickly.

 

A good hitter might be able to hit .400 for 50 games, but he will regress to .300 by the end of the season.

 

But a team who beats their Pythagorean record by 5 games might very well do it again the following season.

 

The correction factor seems weak to me. That is my real point.

 

I'm guessing the reason here is due to sample size. When you say that a coin flip corrects itself to 50% very quickly, I don't know what you consider 'very quickly'. I know that in 100 coin flips, very rarely do the results come out 50-50. In consecutive tests of 100 coin flips, it's not at all unusual to have +8 heads twice in a row.

 

It's a team's lopsided record in one run (or two run) games that often skews their Pythagorean W-L. Even though a season is 162 games, how many one run games does each team play in a season?

 

I get what you're saying about the correction factor. It might take longer to correct itself than we'd like. I just don't like people saying or hinting that we are not a good team or that we don't have a good manager because we can't win the close games.

[Bellhorn04]

I'm guessing the reason here is due to sample size. When you say that a coin flip corrects itself to 50% very quickly, I don't know what you consider 'very quickly'. I know that in 100 coin flips, very rarely do the results come out 50-50. In consecutive tests of 100 coin flips, it's not at all unusual to have +8 heads twice in a row.

 

I tried it a few times when I was a kid. You're right, if you flip 100 times the odds are actually against it coming out 50-50.

 

But it will hover quite closely around 50% if you flip it 100 times or more - that's guaranteed.

[devildavid]

I'm guessing the reason here is due to sample size. When you say that a coin flip corrects itself to 50% very quickly, I don't know what you consider 'very quickly'. I know that in 100 coin flips, very rarely do the results come out 50-50. In consecutive tests of 100 coin flips, it's not at all unusual to have +8 heads twice in a row.

 

It's a team's lopsided record in one run (or two run) games that often skews their Pythagorean W-L. Even though a season is 162 games, how many one run games does each team play in a season?

 

I get what you're saying about the correction factor. It might take longer to correct itself than we'd like. I just don't like people saying or hinting that we are not a good team or that we don't have a good manager because we can't win the close games.

 

Sox are a good team with talent to burn. Losing (and winning) close games seems to be part of the randomness of baseball. In a single game, a bad team can beat a good team in a tight game. The good teams probably win more blowouts.

[Kimmi]

Sox are a good team with talent to burn. Losing (and winning) close games seems to be part of the randomness of baseball. In a single game, a bad team can beat a good team in a tight game. The good teams probably win more blowouts.

 

Exactly. I posted in another thread that if you want to know who the good teams are, look at the records in blowout games, not the records in one run games.

[moonslav59]

To me, it's not about the 16-20 record in one-run games, it's more about our recent trend (last 10 games) of losing low-scoring one run games immediately after a blow out win.

 

It kills me!

[Maxbialystock]

123 Edited September 12, 2016 by Maxbialystock

[moonslav59]

How about the AL East?

 

Teams with a winning percent over .535:

 

4 AL East

2 AL Central

2 NL West

1 AL West

1 NL East

1 NL Central

 

Teams with a winning percent over .549:

3 AL East

1 Everyone else

 

While 4 teams have a better winning percent than the Sox, none of them play in the AL East.

 

 

 

 

[RedSuxlastchance]

See the problem with the red sux is they choke. Not just in 1 run games but it's a thing of habit to you all. To the Orioles? Choke. To the MFY? Choke. You all blow and get ready to get swept go O's¡!

[oldtimer]

See the problem with the red sux is they choke. Not just in 1 run games but it's a thing of habit to you all. To the Orioles? Choke. To the MFY? Choke. You all blow and get ready to get swept go O's¡!

 

We have a troll

[Wade Boggs]

Exactly. I posted in another thread that if you want to know who the good teams are, look at the records in blowout games, not the records in one run games.

 

In the playoffs, you have more one run games. Your facing better pitching. They need to improve on that.

[User Name]

The problem isn't the quality of the pitching they're facing. If anything, the worst part about the one-run losses is that they usually happen against mediocre/sub-par starting pitching.

[moonslav59]

In the playoffs, you have more one run games. Your facing better pitching. They need to improve on that.

 

We've done very well vs top pitching this year. Here's how the Sox have done this year against the top 34 SP'ers by WAR (2014-2016) I'm using the larger sample size than just this year, because if we killed a pitcher this year, we might have knocked him out of the top 34- like Greinke):

W-L ERA/WHIP

Kershaw DNP

Kluber 1-1 4.28/1.46

Scherzer DNP

CSale 1-1 1.29/0.71

J Arrieta DNP

Quintana 1-0 4.73/1.35

Lester DNP

Bumgarner 0-0 1.50/0.83

J Cueto DNP

Keuchel 0-1 12.00/2.17

Greinke 0-1 48.60/6.00

deGrom DNP

Strasburg DNP

C Archer 0-4 6.65/1.75

CHamels DNP

Carrasco 0-0 7.20/1.60

G Cole DNP______

 

Vs top 17: 8-3

 

Tanaka 1-0 2.13/0.79

Verlander 2-0 1.50/0.92

Zimmerman DNP

JFernandez DNP

GGonzalez DNP

CMcHugh 0-1 8.10/2.40

J Lackey DNP

Wainwright DNP

Syndergaard DNP

Samardzjia DNP

Pineda 0-2 4.50/1.50

Hendricks DNP

B Colon DNP

T Ross DNP

Richards DNP

Salazar 0-1 8.31/2.54

S Gray 0-1 17.18/2.73

 

Our record

 

vs the best 34: 12-5

 

I doubt any other team has a better W-L record vs these guys this year.

[moonslav59]

The problem isn't the quality of the pitching they're facing. If anything, the worst part about the one-run losses is that they usually happen against mediocre/sub-par starting pitching.

 

...and right after we blow someone out.

[greenwell]

I love numbers, but if Pythagorean record can be out of whack for as long a time frame as 4 years, it really doesn't mean very much IMO. Sure it works as a general correlation-common sense tells us it must-that's about it.

 

17 wins variation out of 332 total wins is a reasonable statistical variation.

[greenwell]

I tried it a few times when I was a kid. You're right, if you flip 100 times the odds are actually against it coming out 50-50.

 

But it will hover quite closely around 50% if you flip it 100 times or more - that's guaranteed.

 

This is true, just like the yankees being + 17 games over 4 years means they won 52.9% of their games compared to the expected 50.2%.

[Bellhorn04]

17 wins variation out of 332 total wins is a reasonable statistical variation.

 

In most cases, yes, but I'm not sure that applies to a Pythagorean variance.

Edited September 13, 2016 by Bellhorn04

[moonslav59]

In most cases, yet, but I'm not sure that applies to a Pythagorean variance.

 

I guess you'd have to find out how many other teams have ever been +17 or -17 over the same length of time--or thereabouts.

[User Name]

Why not? It's just a formula that measures expected winning percentage via a set of parameters vs actual winning percentage. There's bound to be some background noise, even in relatively larger samples.

[sk7326]

The pythagorean thing is not meant to be a forecasting tool - it provides a context for run differential as an indicator of team strength. I think the "difference" between teams is more useful than the records themselves.

[Kimmi]

...and right after we blow someone out.

 

One of the writers brought up a good point though. Would it make you feel any better if the team won by 3 runs and then lost by one run?

[Kimmi]

In the playoffs, you have more one run games. Your facing better pitching. They need to improve on that.

 

The problem is that there really is no repeatable way to improve on that. You can't beat randomness. The outcome of a one run game often comes down to things like a line drive being hit right at a defender vs a bloop hit falling, or a ball bouncing over the wall for a ground rule double vs staying in the park for a triple.

[moonslav59]

One of the writers brought up a good point though. Would it make you feel any better if the team won by 3 runs and then lost by one run?

 

Actually, in some warped way, yes.

 

The "wasted runs" bug me.

 

I also wouldn't feel as bad winning 16-2 then losing 12-11.

 

It's winning 16-2 then losing 2-1 that drives me nuts!

[Bellhorn04]

Actually, in some warped way, yes.

 

The "wasted runs" bug me.

 

I also wouldn't feel as bad winning 16-2 then losing 12-11.

 

It's winning 16-2 then losing 2-1 that drives me nuts!

 

Winning 16-2 then losing 2-1 is frustrating but it's mostly positive - you just got 2 very well-pitched games in a row. Winning 16-2 then losing 10-9 is much worse.

[oldtimer]

Young is a quality outfielder and his bat is hot now. I suppose he will platoon with Benintendi at left field as soon as Beni gets back. One more positive in a lineup that is a terror right now.

[moonslav59]

Winning 16-2 then losing 2-1 is frustrating but it's mostly positive - you just got 2 very well-pitched games in a row. Winning 16-2 then losing 10-9 is much worse.

 

I know that makes sense, but I can't help but get frustrated with the offense, when we score 16 then 1 and split the two game sample size.

 

This used to happen under the old regime frequently, and I guess it's a by-product of having a high scoring team. You can't expect 5-6 runs every game. The runs are bound to be more randomly dispersed, but over the last two weeks, it's happened 4 times, so maybe it's just hitting a raw nerve that never healed from the "old days" (pre-Henry).