PECOTA projections

[Frobby]

Let me rephrase. You take a pitching staff that was bad last year, lose their best pitcher, and you are going to get a bad projection.

As I pointed out in another thread a few days back, replacing Chen with a starter whose ERA was 2.00 runs higher would cost us 40 runs. This projection is +86. I just can't imagine what else they think is going to occur to cause that kind of increase. Probably some bullpen regression, but do they really expect no improvement from Tillman/Gonzalez/Gausman? Oh well, I'm not going to lose any sleep over it considering the track record here.

[weams]

So they say we will score the fewest runs in the division and also have the worst FRAA (fielding runs) plus our OBP will be worse than last year. Strange, right?

Right. Fools.

[weams]

<blockquote class="twitter-tweet" data-lang="en"><p lang="en" dir="ltr">Projection systems hate Royals. 2 big reasons: 1) Blind spot for effect of defense/pen 2) Don't see luck repeating <a href="https://t.co/GFf05OGeUj">https://t.co/GFf05OGeUj</a></p>— Jonah Keri (@jonahkeri) <a href="

">February 16, 2016</a></blockquote>

<script async src="//platform.twitter.com/widgets.js" charset="utf-8"></script>

[Redskins Rick]

Projection systems hate Royals. 2 big reasons: 1) Blind spot for effect of defense/pen

2) Don't see luck repeating

https://t.co/GFf05OGeUj

— Jonah Keri (@jonahkeri)

February 16, 2016

<script async="" src="//platform.twitter.com/widgets.js" charset="utf-8"></script>

Any sport, its always harder for the champion to repeat, since everybody knows you was king of the hill last year, and they are gunning for you, this year.

[El Gordo]

The most important thing I got from any of this was:" Let's admit it: I could pretty much list all the remaining American League teams at 81-81. But that wouldn't be any fun!"

Puts all this projection stuff in perspective IMO.

[BohKnowsBmore]

As I pointed out in another thread a few days back, replacing Chen with a starter whose ERA was 2.00 runs higher would cost us 40 runs. This projection is +86. I just can't imagine what else they think is going to occur to cause that kind of increase. Probably some bullpen regression, but do they really expect no improvement from Tillman/Gonzalez/Gausman? Oh well, I'm not going to lose any sleep over it considering the track record here.

We also shed 11 starts of 7.06 ERA that represented a -1.5 rWAR. In theory, just replacing Norris's contributions with some replacement level production should lead to some improvement in the rotation (partially offsetting the loss of Chen).

[BohKnowsBmore]

<blockquote class="twitter-tweet" data-lang="en"><p lang="en" dir="ltr">Projection systems hate Royals. 2 big reasons: 1) Blind spot for effect of defense/pen 2) Don't see luck repeating <a href="https://t.co/GFf05OGeUj">https://t.co/GFf05OGeUj</a></p>— Jonah Keri (@jonahkeri) <a href="

">February 16, 2016</a></blockquote>

<script async src="//platform.twitter.com/widgets.js" charset="utf-8"></script>

I have said this in the past. Recently I have said the same thing about KenPom for NCAA Bball. Maybe they should refer to variance/randomness/etc. in the models and results as something other than "luck." I feel like even among those who understand that it's not supposed to denigrate teams in the "oh, you were just lucky!" sense (well, at least by most who use it), it's still a fairly loaded or aggressive term. Just my $0.02 on that.

[Can_of_corn]

If healthy, the floor of this team is .500, IMO. Anything in the 70's feels patently ridiculous and indefensible without significant injuries.

Add in Gallardo and Fowler to the projections and the floor is still .500, but the ceiling is 90.

There. Put that projection in your pipe and smoke it.

I don't see how you can take an 81 win team, with a pythag of 83 wins, who won their last six games against depleted rosters, that lost their best starter, and say that a projection they win 79 games the following year is indefensible.

[Can_of_corn]

Well, because I'm a poster on an internet message board, mainly.

Fair enough.

[Redskins Rick]

I don't see how you can take an 81 win team, with a pythag of 83 wins, who won their last six games against depleted rosters, that lost their best starter, and say that a projection they win 79 games the following year is indefensible.

I like Dan's response.

But, this also doesn't take into consideration, how many winnable games they blew, which was above average for them.

So scale that back down into some method of reasonableness.

I guess it just shows you the very fine line between 20 games over .500 and .500 team.

[Can_of_corn]

I like Dan's response.

But, this also doesn't take into consideration, how many winnable games they blew, which was above average for them.

So scale that back down into some method of reasonableness.

I guess it just shows you the very fine line between 20 games over .500 and .500 team.

Actually I would say that by mentioning their pythag last year I did take into account the winnable games they blew.

I do agree that teams may win or lose a drastically different number of games then their talent level would indicate.

[25 Nuggets]

Actually I would say that by mentioning their pythag last year I did take into account the winnable games they blew.

I do agree that teams may win or lose a drastically different number of games then their talent level would indicate.

That's what I'm hanging my hat on. Variance.

On paper I think the ZiPS projection of 83 wins sounds about right. That's without Gallardo and Fowler. I'm guessing those two bring the projection up to maybe 86 wins. That's good enough to roll the dice with.

[TradeAngelos]

If healthy, the floor of this team is .500, IMO. Anything in the 70's feels patently ridiculous and indefensible without significant injuries.

Add in Gallardo and Fowler to the projections and the floor is still .500, but the ceiling is 90.

There. Put that projection in your pipe and smoke it.

Well I don't know about "anything in the 70's" when that could easily be 79 wins or 71 wins. I don't think anyone would be shocked at 79. Or 78...or 77...or 76....

72 wins...71....70? I think that is an area where people would be shocked, and significant injuries would have to come into play.

But I think the higher win totals over/under 82 are much more likely than the lower ones.

[tntoriole]

It does seem like quite the annual exercise in masochism for us fans (who admittedly are not necessarily objective or even rational as it relates to the Os). Every year since 2012, these "prediction" methods have put the Orioles last or near last in their division, usually under 80 wins, and then every year, the predictions have turned out flawed at best and wrong at other times.

I remember as a kid getting the Sporting News predictions magazine in the mail where it would predict the standings for the coming year- no fancy methods, no analytics, no peripherals, no Bill James wannabes...just sports writers and baseball types coming up with consensus opinions. It would be interesting to go back to those archives and compare the accuracy of the predictions for team W-L records from 1955-1970 to the predictions using these methods from 2000-2015.

When they do get it wrong, then they start using terms like "luck". Rather than just acknowledge that their methods are no more accurate than, say, posters on the Internet, for example. A baseball season is a complex multivariable activity involving a group of 25 world-class athlete human beings engaged in the thousands of performance tasks of the unique and individual baseball season. I do read the projections each year with some interest but I put little stock in their conclusions, even if those were to rate the Orioles at the highest of levels.

[Frobby]

Actually I would say that by mentioning their pythag last year I did take into account the winnable games they blew.

I do agree that teams may win or lose a drastically different number of games then their talent level would indicate.

At about the all-star break, the O's had a terrible record in one-run games. That reversed in the second half and they ended up 25-26. So, I really don't see that they lost a disproportionate number of winnable games. They were a so-so team last year that had poor starting pitching and an offense that was decent but prone to cold spells.

I could easily imagine the Orioles finishing under .500 this year. But not 72-90 as PECOTA projects.