BP: Baltimore's Annual Surprise Party

[atomic]

In case anyone has forgotten:

http://forum.orioleshangout.com/forums/showthread.php/151731-PECOTA-projections/page7?highlight=pecota

The Rays just have to go 59 and 28 the rest of the way to bail them out. I don't see how they think Tampa was going to outscore us. Have they seen their line-up?

[LookinUp]

Chris Davis has an .853 OPS, is on pace for ~40 home runs, has played plus defense and in total is on pace for 4.5-5 WAR.

He's clearly a massive disappointment. Makes me wonder who even wrote the article. It's not hard to figure out from the stats what we're doing. We're outscoring teams and beating them later in games with our great bullpen. It will continue in the second half until we run into the teams against which we can't score 5+ runs a game.

[LookitsPuck]

This is just asinine. I dunno how anyone with any sense could say we'd be in last place in the AL East, and especially have the worst record in the AL. That's just dumb. I could understand saying we're a 75-80 win team, but worst record in the AL? Come on. We have one of the best offenses in the league and one of the top 'pens in the league. That alone is enough to keep those two things from happening even if the wheels fell off somewhat like last year. IMO this teams floor is 75 wins, and it's ceiling is somewhere around 90.

I really thought this team was going to be mediocre at best. I thought the rotation was going to be historically bad (because I still don't buy Wright, Ubaldo and Gallardo), but Tillman and Gausman have really buoyed this staff and Wilson has been serviceable. The offense has hit above expectations. That's really what is carrying this team. Esssentially 2 out of 5 turns through the rotation you'll get a good pitching performance (although Tillman is regressing) and 1 time you'll get a solid start (Wilson)...the other 2 I think are going to be clunkers...but if 60% of the time you can remain competitive then the other 40% of the time can be up to pitching matchups and offensive onslaughts...which I'd imagine can go in your favour every now and then. Probably not a winning proposition come playoffs, but there's enough variance there for it to *maybe* work.

I still am not looking forward to 2018.

[weams]

I think some of the problem is his defense this year.

His defense is fine. The metric might be off again.

[weams]

I really thought this team was going to be mediocre at best. I thought the rotation was going to be historically bad (because I still don't buy Wright, Ubaldo and Gallardo), but Tillman and Gausman have really buoyed this staff and Wilson has been serviceable. The offense has hit above expectations. That's really what is carrying this team. Esssentially 2 out of 5 turns through the rotation you'll get a good pitching performance (although Tillman is regressing) and 1 time you'll get a solid start (Wilson)...the other 2 I think are going to be clunkers...but if 60% of the time you can remain competitive then the other 40% of the time can be up to pitching matchups and offensive onslaughts...which I'd imagine can go in your favour every now and then. Probably not a winning proposition come playoffs, but there's enough variance there for it to *maybe* work.

I still am not looking forward to 2018.

I don't care about two years from now. I don't care about two years from now...

[CanadianOriole]

I'll own it - I didn't like any of Tampa, NY, or Baltimore's teams coming into this season and pegged those 3 to all be sub .500. I wouldn't have picked us for last - but I thought we'd be in the mix for last. Machado, Schoop, Wieters, and Trumbo are all sporting OPSs about 100 points higher than I would have projected. I didn't have Trumbo leading the league in home runs. I didn't have Chris Tillman at 10-1 at this stage of the season. Our pitching as a whole has been middle of the pack - I expected it to be worse than that. I expected Gausman to be the only above average starter in our rotation. I follow this team extremely closely as we all do, and I've been doing it a long time - and I was wrong all over the place about this team so far. Predicting what's going to happen in one individual baseball season is very, very hard.

[weams]

Chris Davis has an .853 OPS, is on pace for ~40 home runs, has played plus defense and in total is on pace for 4.5-5 WAR.

He's clearly a massive disappointment. Makes me wonder who even wrote the article. It's not hard to figure out from the stats what we're doing. We're outscoring teams and beating them later in games with our great bullpen. It will continue in the second half until we run into the teams against which we can't score 5+ runs a game.

Of course. HE is the Baltimore Orioles.

[LookinUp]

Of course. HE is the Baltimore Orioles.

I should add that I didn't read the article, just the OP. I think the post above yours points out the 7 players who are outperforming the stats. That's where some of the delta is. The other delta really is associated with Buck, IMO, and you can't measure that. The only comparable managers, IMO, are Madden and Girardi.

[psagawa]

The predictive power of any of these projection models is low - the standard deviation of prediction vs. actual wins is something like 10%. The models have been particularly bad at predicting the O's (also the Royals) and the poor performance of the model has persisted for the past 4 years. The big mistake that sports writers make is assuming that unexplained variance in the model is "luck", as though the model was perfect. As a stats guy, I'd say that Duquette has found something unexplained by the model that he can exploit. Maybe it is relief pitching or power or Buck, who knows but Duq, but it is real.

[weams]

The predictive power of any of these projection models is low - the standard deviation of prediction vs. actual wins is something like 10%. The models have been particularly bad at predicting the O's (also the Royals) and the poor performance of the model has persisted for the past 4 years. The big mistake that sports writers make is assuming that unexplained variance in the model is "luck", as though the model was perfect. As a stats guy, I'd say that Duquette has found something unexplained by the model that he can exploit. Maybe it is relief pitching or power or Buck, who knows but Duq, but it is real.

I would say that is most accurate.

[Frobby]

I do believe, by the way, that our offense is at an apex right now and will not score 5.03 runs/game or post an .802 OPS over 162 games. It's just the ebbs and flows of the season. We scored the fewest runs/game in the AL in May, and are the highest in June. The bats will cool off again eventually. I'm just enjoying the ride while we're hot.

[SpinJr]

As a stats guy, I'd say that Duquette has found something unexplained by the model that he can exploit. Maybe it is relief pitching or power or Buck, who knows but Duq, but it is real.

CoC said something similar, but what if there is no data that only he has? What if it just simply something that is unquantifiable. The X factor so to speak. I'm not convinced Dan has some special data only he has discovered.

Agree 100% in whatever it is, quantifiable or not, is REAL.

[Why Not?]

It is good they don't understand how it works. That way the O's can get guys that fit their formula for victory without other teams knowing how it is done.

I agree. Though I think the point of the article was that the Orioles don't really have a "formula" other than adding value through late free agent additions.

The only consistent thing I see in the Orioles' roster-building is the willingness to make a LOT of small incremental improvements that eventually lead to a winning team.

[SilentJames]

The Orioles' biggest weapons don't always show up reliably on projections: defense and bullpen.

There was an article somewhere that went into how Bud Norris was a perfect example of how sneaky the Orioles are. He didn't have a lot of QS, by definition but he very rarely let any game get out of hand. So the Orioles homer-fueled offense could feast on opponents' bullpens and win a lot of games late and close.

So the Orioles weak SPs really just needed to hold the fort for their six innings and the defense and offense would keep them in so many games. The pen comes in and shuts them down.

Sent from my iPhone using Tapatalk

[birrman54]

Their decision to replace the steadying force of Wei-Yen Chen with questionable in-house options and a bargain-bin Yovani Gallardo appears, to this point, unfortunate.

This makes no sense.

Gallardo we signed for 2y/22M. Chen was signed for 90M.

Chen has a 5.00 era in 15 starts so far this year. If we'd signed him instead wouldn't he be getting lumped in with the "paying too much for our own free agents" bin?

Also confusing how they're not aware that Reimold has been a effective (when healthy) role playing outfielder most of his career and his 105 wRC+ is basically in line with his career 102 mark.

I think often times people are too critical of PECOTA because they compare it to specific counter-examples, rather than the general predictive nature ability - but article's like this don't even make the right case for why projections can be reasonable and still wrong.