How bad can you be Part two

To build my rating system, I made a lot of approximations about the value of running, of arm defense, of catchers' arm, etc, etc. It's basically the same idea than Dean Carrano's formula in his OFFENSE vs DEFENSE pdf, but using different formulas, and taking EVERYTHING on a card into account--running, defense, gbC. But as I said, I made a lot of approximations, so when I see discrepancies between my system and SOM pricing list, I can't always figure out whether it is SOM or my approximations who are offline.

How exactly did you make these "approximations" about the values of "runnning, arm defense, catcher's arm, etc.", and how did you come to those approximations? In other words, how did you respectively value each aspect of an SOM card below and above other aspects, and how did you assign numbers to those values.

Obviously, there have been years of debate over what aspects of a card/parts of the SOM game are more important than others, and what each is worth. If you've actually figured it out, you could put together a ratings guide more popular than JoetheJet's

Marc,

How about a link to your team?

Regarding arm strength, which I pretty much ignore, I have found Harper (-5) to be of decent value. But I too have also used Crisp (+2) with some pretty good success, esp in hitters parks. I can't say I would ever have to or not have to use someone based on their arm, catchers included.

Regarding Utley, I've used him, but only when I can't get a few others I like a bit more. I don't remember him standing out, but I'll go back and check when I get a chance.

Glad to see you back on the boards -

Whenever you see a team that wins a ring using virtually no guys you would ever use, it really makes you think hard about how overrated card-reading just might be.

That could be.
Is it also possible that manger might be better at card reading? Perhaps such a team should make you think hard about what values you are assigning to how you read those cards.

My own experience with trying to count everything on a card and arrive at an end value is how to value various ratings. It is fairly easy and straightforward to count the types of hits. Easy to use some saber runs created formula to calculate the value of those. Easy to calculate ballpark effects. I even at one time had a formula for calculating the value of a gb()C. But how much is an A bunt rating worth verses a B rating with respect to runs created? Throwing arms? Run and steal ratings?

Could it be that some of those who are able to win a ring with cards I would never use just understand how to value all these ratings better? Could it be they have a better understanding of how all the individual and team settings impact a team player and team?

In other words just that someone wins with cards we would not use does not invalidate card reading.

To build my rating system, I made a lot of approximations about the value of running, of arm defense, of catchers' arm, etc, etc.

You should start a thread and list some of your formulas. I for one would greatly enjoy a conversation on the how and why of the formulas.

and taking EVERYTHING on a card into account--running, defense, gbC

I would especially enjoy this part of the conversation.

Since Valen re-posted my exact questions, I'll re-post and elaborate on mine so you don't have to go back:

To build my rating system, I made a lot of approximations about the value of running, of arm defense, of catchers' arm, etc, etc. It's basically the same idea than Dean Carrano's formula in his OFFENSE vs DEFENSE pdf, but using different formulas, and taking EVERYTHING on a card into account--running, defense, gbC

How exactly did you make these "approximations" about the values of "runnning, arm defense, catcher's arm, etc.", and how did you come to those approximations? In other words, how did you respectively value each aspect of an SOM card below and above other aspects, and how did you assign numbers to those values?

I'm also interested if and how you alter those values for each park. Presumably, arm defense, running and speed will have greater value in a pitcher's park than in a hitter's park. So, I'm interested if you make adjustments for such shifts in your formula.

I was also interested about gbC chances not too long ago. It turns out there's not too much of a difference between cards: How to find hitters who make productive outs

I am confused, because on that older thread in the general strategy that Wavy links to it someone wrote, (in reference to ground ball C chances on hitter cards) that most hitters had a maximum of 6 chances and;

Check out ..... Over the 6 limit but not enough to matter.

Sheffield 03
Ruppert Jones 77
Dolph Camilli 38
Tony Gwynn 94
Richie Zisk 77

Pujols 08
Milt Smith 55
Dick Allen 64
Albert Belle 95
McGwire 92

But I can't find ANY ground ball C chances on Richie Zisk's 1977.? My reading comprehension or understanding of what was being discusses faltered me..... This is an interesting topic as well!

ps. I did not mean to hijack the thread, ignore my question and I will not be offended. Sorry for the intrusion!

Could it be that some of those who are able to win a ring with cards I would never use just understand how to value all these ratings better? Could it be they have a better understanding of how all the individual and team settings impact a team player and team?

In other words just that someone wins with cards we would not use does not invalidate card reading.

Sure, somebody can always have more knowledge I guess. I'm not sure if how you understood what I was saying was what I meant to be expressing, so let me try and rephrase...

Let's use a model which consists only of real life MLB RC-27. In this model, $0.53 Delmon Young and $6.45 Matt Holliday both share a 6.20 RC-27. Now RC-27 takes into account some fairly relevant data (obp, slg, etc), I think we agree on that. And you do not have to be a math genius to intuitively understand that there's inherent value in that Young's RC-27 is 100% of Holliday's at 8% of the salary cost. There's a lot of what I would call "margin of error" built in there. So I have made a significant discovery, right?! I can have a nine man lineup of Young's costing $4.77 versus a nine man lineup of Holliday's costing my opponent $58.05, theoretically producing equal RC-27. And even if I'm really, really wrong, my $75 mil pitching staff will be taking on his $20 mil staff!

Why does this not work in reality (Strat reality, not real world reality)?

My ability to recognize value does not necessarily equate to my ability to field a team who ultimately wins a ring. If I count points in the ratings book and try and maximize points to salary, I will come up with many, many efficient lineups. Just because someone else can do the same and ultimately win a ring, it doesn't necessarily follow that they were a better "card reader". It only means that in this particular instance, his selections worked out better than mine (ie, he won the ring). In fact, in a league of 12 experienced managers, a strict points per salary model might even indicate the manager with the least value added (points per salary) could still win the ring.

Now, let's go one step further and add clutch. On my last team, I had Beltran, who has 11 points of clutch on each side (very high). He was 4 for 4 in clutch opportunities - awesome! But did I originally draft him so he could contribute 4 extra awesome at-bats in a season? Isn't that a little too much can't see the forest for the trees stuff? How far in do we have to go to say we're a better "card reader"?

Maybe Strat already has the ultimate most perfect way to account for each and every variable in baseball and everything is priced in the most efficient mathematical fashion into each card. So building a better mousetrap might not be adding any value whatsoever. Then it might follow that the ability to ferret out the clutch, and the bunts and the arm, and the rest of the "noise" from the salary is what card reading is all about. Making things less complicated, not more.

Anyway, way too long. My point is, what we call "card reading" might be too difficult to differentiate between 12 equally good players. So I'm not saying "invalidates" is the right word, maybe "effectively cancels each other out" is a better way to say it.

One can almost completely ignore out-types, base running, stolen bases and still be a dominant Strat manager. I never count gbc's, or build teams around stolen bases as a primary organizing factor. Count offensive and defensive times on base and total bases on offense and defense and you are 99% of where you need to be. The difference between good and bad managers on here has much more to do with bad managers making errors on pricing-to-bases earned(obp, slug) evaluations than in failing to compute the minute value of productive outs and the like. Don't sweat the small stuff is the best advice, particularly for struggling or new managers.

Now, do I look at a player's speed, stolen base success rate, and gb "a"'s when setting my lineup order? Of course. But it's not something to obsess over when drafting a team at the outset. on base chances on offense and defense, slugging numbers, park suitability, are the things to sweat. And a good ratings guide can really shorten the time spent in crunching the important things.

The difference between good and bad managers on here has much more to do with bad managers making errors on pricing-to-bases earned(obp, slug) evaluations than in failing to compute the minute value of productive outs and the like.

I like that. I would even go so far as to split the quote in half: "The difference between good and bad managers on here has much more to do with bad managers making errors."