Using Diamond Dope

[PotKettleBlack]

Dean beat me to it, but I put it up as well:
https://docs.google.com/viewer?a=v&pid=explorer&chrome=true&srcid=0B6PjcDHBxnNPYzllYjhmM2YtMzVhNS00ODExLTljZWYtZWU5MTIxM2Y0ZWYw&hl=en_US

[PotKettleBlack]

[quote:8e31b1dc69="motherscratcher"]When you look at actuals do you look at the results for all of the seasons played? Or do you limit it to the ballpark you are in?

(Also, where does all of the data come from for the actuals?)[/quote:8e31b1dc69]

Actuals with a grain of salt. Definitely at cap level (cap level skews context a lot). I look at hitters parks if I am curious about parks within a cap. See how others are using someone, see what their thinking was a bit, see what I might expect.

Context matters (central point in all things baseball) and even in the same park at the same cap, the oppositions' parks may be skewed considerably from league to league.

A second word of caution with actuals. Fairly robust on the guys who get used all the time. Possibly very skewed on the guys who don't. Small sample sizes, damn them.

A third word of caution. Win %age. Interesting to look at, but very easy to overstate. I like to consider them a little, as a value metric on the larger dollar players. The lower the dollar figure, the further removed from a key role on a winning team the player probably is.

The actual data comes from actual leagues. No live draft leagues. This leads to another skewing factor. Theme leagues are given equal weight. As are limited set leagues. As are 24 team leagues. Something to consider, as a league that limits your set can skew things quite a bit.

[nels52]

First off, Diamond Dope is amazing (thank you Agabriel)

Secondly, never understate Defense. While there may be no way of quantifying how much it helps your team just know that it does.

Defense wins championships, games, series' and divisions.

The difference between a 3 and a 2 in let's say LF is astronomical.


In my opinion, the biggest deciding factor in Strat is lineup efficiency. While loading your lineup with studs may be tempting, it doesn't make much sense to do so if you've got huge OBP guys like Ted Williams if you don't have anyone to drive him in.

For example: If you're building a team around someone like Ruth or Bonds there is a certain obligation to surround them with high OBP men in front of them and big RBI guys behind them.

This is fairly obvious in the top of the order but not so much in the 5, 6, 7 spots in the order.

[size=18:d4fa2ef328]Having too much OBP in the 6 spot in the order can be one of the most deadly things to a team.[/size:d4fa2ef328]

[PotKettleBlack]

[quote:45870b96ce="nels52"]First off, Diamond Dope is amazing (thank you Agabriel)

Secondly, never understate Defense. While there may be no way of quantifying how much it helps your team just know that it does.

Defense wins championships, games, series' and divisions.

The difference between a 3 and a 2 in let's say LF is astronomical.


In my opinion, the biggest deciding factor in Strat is lineup efficiency. While loading your lineup with studs may be tempting, it doesn't make much sense to do so if you've got huge OBP guys like Ted Williams if you don't have anyone to drive him in.

For example: If you're building a team around someone like Ruth or Bonds there is a certain obligation to surround them with high OBP men in front of them and big RBI guys behind them.

This is fairly obvious in the top of the order but not so much in the 5, 6, 7 spots in the order.

[size=18:45870b96ce]Having too much OBP in the 6 spot in the order can be one of the most deadly things to a team.[/size:45870b96ce][/quote:45870b96ce]

Really, it depends on the style of team. A high OBP, base stealer in the 6 hole, with a bunt/H&R guy 7th and a +clutch guy 8th in front of the pitcher can actually turn some runs in cheaply, allowing more investment into bats 1-5. Especially if 6-7-8 are good gloves.

[Munich_Man]

[quote:f8966fca15="River Field View"]
Honestly, though, I usually only look at the actuals if
I think I am missing something in the what I call the DD Projections (the
card reader page).[/quote:f8966fca15]

Funny, I ONLY look at the actuals. I'm much more interested in how cards actually perform in real leagues than in arcane projections of how they SHOULD perform (and frequently don't).

[doug_tucker10]

[quote:3814b8c0f8="Munich_Man"][quote:3814b8c0f8="River Field View"]
Honestly, though, I usually only look at the actuals if
I think I am missing something in the what I call the DD Projections (the
card reader page).[/quote:3814b8c0f8]

Funny, I ONLY look at the actuals. I'm much more interested in how cards actually perform in real leagues than in arcane projections of how they SHOULD perform (and frequently don't).[/quote:3814b8c0f8]


Same here ....

[216 Stitches]

[quote:486268c00b="Munich_Man"]
Funny, I ONLY look at the actuals. I'm much more interested in how cards actually perform in real leagues ....[/quote:486268c00b]

That's certainly the better advice for a newcomer.

And there is no question that the actuals drive the popularity of specific
cards.

On the other hand, the projections and the actuals shouldn't be
in disagreement without a reason. And understanding why they
differ gives significant insight into more subtle strategies.

Since I already get a feeling how many of the most commonly used
cards are actually performing in my own leagues, the actuals are kind
of redundant for me. I know the statistical size in the DD actuals is
much larger, but in the leagues I play in I do have more specific
context (exact stadium, division makeup) to go by.

[ADRIANGABRIEL]

[quote:6c6674c1d2="DeanTSC"]Looks like there is no longer a "blog" function, plus the site that hosted the article is down anyway :(

For now, I guess I'll put it [url=http://www.box.net/shared/fifc1lhyn3axupgln540]here[/url].[/quote:6c6674c1d2]
Dean has quantified defense and put it in the form of NERP--New Estimated Runs Produced, a linear-weights approach similar to Batting Runs.

The Card Reader calculates NERP and also factors gbA. All you have to do is add the numbers together.

I never have worked these defensive values into the formulas. I am lazy.

[motherscratcher]

The last team I built I relied heavily on both the actuals and the card reader. Basically, I looked at the actuals to see which players seemed to perform well and which cards where used the most. then I would look at those cards in the card reader to see how they seemed to stack up against each other and perform in my park.

All things being equal, the card should perform like the card reader says. It's just math. Over the long term, dice whould not defy the laws of probability or produce a rip in the space time continuum. If a card under/overperforms, there should be an explanation as to why that happened. Not that I would ever be able to figure it out.

But just relying on the actuals seems like folly to me. There are too many things we don't know to explain a performance, and when we whittle those things down with the cap and/or the park, the sample becomes very small. Often the card you are interested in hasn't been used at all.

So I spent hours meticulously building a team for griffith '57. Then I blew the whole damn thing up after half the team was lost in the draft.

That meticulously constructed juggernaut got swept last night (third sweep on the young season) and is currently getting it's *** kicked from foul line to foul line and beyond center field.

Good times.

[216 Stitches]

[quote:9013bab086="motherscratcher"]
That meticulously constructed juggernaut got swept last night (third sweep on the young season) and is currently getting it's *** kicked from foul line to foul line and beyond center field.
[/quote:9013bab086]

After my 8th SOM League, I decided to build a team just on DD projections.

Of the 80 teams that I have finished, only one team has had a worse
record than that team.

I think DD is a gold mine, but it has a Clint Eastwood aspect to it:
Man has gotta know his limitations.