R.I.P. Moneyball Part II –The UNDERVALUED STAT!

An article by posted on October 17, 2011

Having explained more precisely what the central GOAL of Moneyball is (Save Money) I thought it might be good to talk about what it actually says it did and why it did what it did from the Statistical Analysis point of view.

Moneyball, as was stated previously, used Statistical analysis to find an UNDERVALUED METRIC that was then used to hunt for and create a MASTER LIST of players who exhibited that quality for consideration of acquisition.

How did they arrive at OBP as the undervalued metric and how would that work today?

Well the book leads us to believe that Sabermetrics were used to list and rank players and to determine which stat was prevalent in good players AND cheap players that was being overlooked by the rest of the league.

But did that really happen? Did it really show that OBP was the undervalued stat or did it by statistical bias automatically select OBP as the undervalued stat since Sabermetrics were used and invented to ILLUSTRATE the importance of OB in the game of baseball?

You see by using Sabermetrics as your analytical model you PRE-SELECTED OBP as valuable merely because Sabermetrics values that more than any other event and didn’t really let the analysis of numbers make the decision for you.

You would come up with OBP as the UNDERVALUED VALUE stat even in years when it was not.

If you didn’t construct the statistical model in a way that made OBP valuable before any comparison was made it would not have valued OBP in the first place and therefore you would not have come to the conclusion that OBP was UNDERVALUED!

Moneyball used an OB-centric statistical model and came up with OBP as the UNDERVALUED METRIC. WOW, What a SHOCK! They looked for something specific and then FOUND IT! Eureka moment or preselected destiny because they used a metric that favored the result before the result was even seen?

IE: If you ran a statistical analysis model that favored HRs above any other event you would come up with a list of GOOD and CHEAP HR hitters, and you might be left thinking HRs are the UNDERVALUED STAT, but would you be right? Would it mean that HRs are undervalued? No – what you did was lent value to something before any player comparison was made. You FOUND what you BIASED your search parameters to look for. HRs are a poor example granted because we all KNOW they are valued, but it does make the point that if you bias your stat model towards one event, it will only work if your biased model correctly ascertained the undervalued assets from the outset. Garbage in, Garbage out.

Now OB was undervalued at the time because OBP while well known, was not given the importance it is today thanks to the book Moneyball. Most teams all look at OBP now and it is no longer undervalued – it hasn’e been for quite a few years.

Sabermetrics were new, not widely used and therefore no majority of teams would come up with the same MASTER LIST as Oakland meaning they would not be FAVORING OBP and as a result OBP would be undervalued.

This is how MONEYBALL came to find that particular market inefficiency.

Would it work today?

What is the most UNDERVALUED STAT in today’s market?

Can Sabermetrics tell you what is undervalued these days?

Or will Sabermetrics come up with the SAME EXACT ANSWER as before because it values OB more than anything else?

It is believed that Sabermetrics actually found the UNDERVALUED stat but the truth is the guy who PICKED Sabermetrics as his statistical model picked OBP as the undervalued metric before any comparison, chart or calculation was made!

And it is for this reason none of the people who have a strong belief in Sabermetrics have been able to ascertain what the new UNDERVALUED metric is.

ASK them and they will not be able to tell you. Any of you sabers care to list a couple of undervalued metrics right now?

They will cite other metrics made BASED on the same OBP biased approach, leading to the same results.

They need to come up with a new statistical model that will not value one stat over another and not bias the results in favor of OBP. Otherwise just use OBP and trash the other stuff that doesn’t really lead you to anything undervalued assets.

Sabermetrics was primarily a form of statistical analysis that allowed OBP to be seen as underused and undervalued. That is all behind us now and what we now need is to discover that NEW undervalued Metric.

A NEW STATISTICAL APPROACH IS REQUIRED if that Moneyball UNDERVALUED approach is going to ever work again!

That can only occur by someone finding an event that is overlooked and creating a metric to find it.

If Sabermetrics is used by all teams then Sabermetrics will not be able to come up with an UNDERVALUED STAT because EVERYONE will be valuing it!

SO for MONEYBALL to succeed it has to abandon the Sabermetrics it used when no one knew about it, and it must invent a NEW Statistical model that no one else uses in order to get back its edge and find what no one else is looking for and placing VALUE on!

And if that’s not done?

R.I.P. Moneyball.

This Fan Shot was submitted by Mike (Metsie). Have something you want to say about the Mets? Share your opinions with over eleven-thousand Mets fans who read this site daily. Send your Fan Shot to GetMetsmerized@aol.com.