[ScreaminPain]

I’ve been working on some correlation situations and have run across some discrepancies’ in data from site to site. All in all, data error has been minimal, but I’ve stumbled upon this one item that has me baffled. It appears that STATFOX has some arrangement allowing SCORESANDODDS to use some data. That’s fine, but when perusing the Power rating data from these sources I noticed some strange differences.

This is the quote from STATFOX site as to how to apply their power ratings.

To calculate a hypothetical line, divide the higher rating by the lower rating. Multiply this result by 100 and add 14 to the result for home field advantage.

OK!;.....that doesn’t seem too hard. Today, the ANGELS are rated 100 and TEXAS is rated 114. Game is being played in Texas. So, dividing 114 by 100, then multiplying by 100, then adding 14 for home field gives me a total of 128. So that makes TEXAS a -128 favorite in todays matchup according to the STATFOX ratings system.

But wait there’s more…..If you go on the SCORESANDODDS sight, scroll to the very bottom and select “baseball matchups” a new screen opens and you’ll see a listing of all of todays games. Next to each game is a clickable link titled “report” which brings up a detailed data sheet with a heading of STATFOX. I’m assuming the sites have some sort of agreement to display each others data, but that’s not the point here. Near the top there is a section titled “Power Rating Line”.

Here is the quote on directions for interpreting the line: The power rating system is based on recent game results. The section should be used to compare the relative strength of the two teams involved. A power line is calculated using both teams ratings and home field advantage. Where the power line differs significantly from the current line, the team with the edge is indicated.

In a section accompanying the data is a place for the teams name, pitcher, estimated line and EDGE (if any).
OK, this is even easier…….It appears all the “work” has been done for me. Simply read the data and see how it compares to my line. To make it even easier, if the EDGE column has a star in it, that means they have deemed that team has a value vs. the posted line.

Here’s the problem. In this section, they have TEXAS as a -186 favorite, WITH A STAR in the EDGE column. How can TEXAS be a -186 fav. On SCORESANDODDS site and -128 fav. On STATFOX site, since their BOTH USING THE SAME RATINGS!!

I can understand small differences in data from site to site, but 60 points on a betting line is not acceptable by any means. And the trouble is: WHICH LINE IS CORRECT? Or are either of them realiable? I seriously doubt the ratings are derived using only the posted directions, especially in the case of SCORESANDODDS data, but if additional data is included in the final analysis, WHY ISN’T THAT MENTIONED? In moving forward, I"m concerned about the validity of any data purportedly used to create measureable value.

[RickySteve]

I would speculate that the former system is a general power rating, the latter adjusted for starters, and both garbage.

[uva3021]

do they have power rating for starters?

[ScreaminPain]

do they have power rating for starters?

NO! I'm certain that is part of the reason for the differences, but shouldn't the site explain that the ratings will differ depending on starters? Each site states the directions for using the ratings (quoted in original post) and they don't make any mention of needing additional data.

I personally don't use any of these numbers and only encountered the differences while compiling ratings for possible correlations.