how to quantify qualitative Data.

[TheDominican]

Let's say that someone has a way to find out information that affects a result that is not reflected in the data i.e: a certain player is having marital problems, or athlete x is using a new product that gives him an edge, etc.

How can one give proper value to that information, how can one asses the impact on such information in the final result of a game?

[lamichaeljames]

tough question - i want to hear an answer to this as well. I guess it would have to depend on the specific sport, due an individual having a greater impact on the outcome of said sport.

[TheDominican]

You are right, I suppose the position of the player matters too, a starting pitcher in baseball has a lot more impact than any other position for instance.

[magyarsvensk]

Assign an optimum value to each of the possibilities that can be added to or subtracted from a baseline prediction (such as closing odds or spreads). So if you were magically able to determine the status of every player's marriage at any given time during the season for a large enough sample, you could measure the effect that had on the outcome of the game over say 10,000 games.

A more realistic example might be the handedness of the starting pitcher and the other team's lineup. That data should be more or less available.

[High3rEl3m3nt]

@TheDominican,

That is not qualitative data. That is quantitative data that has not been collected.

In both of your examples, it is damn near impossible to know the relationship between those variables
and the outcome in focus, because there are a shit-ton of independent variables that you can't control
or are arguably more important in affecting the outcome.

[T4TRUTH]

But, and there is always a but! I have had an experience once upon a time to gain some locker room access to NBA games and yes it is a huge deal.
Because quite simply knowing a particular player who is 80 to 90 % three point shooter, having had a night of bad stomach due to food and an argument with the misses,caused a terrible shooting night and just filling up space on the floor.
Having witnessed that yes there is massive amount of unknown data that causes change to final outcome of games.
Next point along same lines. and to me much more important as the influence and edge is too great. WHAT ?????is the status of the referees. what is going on in their personal life. much greater amount of data too be sought.

[High3rEl3m3nt]

You can directly observe something and make an inference based upon your direct observations, which is qualitative.
The key is to not overvalue what you've observed or to ignore as to whether the market has already corrected itself.
These days, this type of scenario rarely goes unnoticed with social media.

[brettd]

Let's say that someone has a way to find out information that affects a result that is not reflected in the data i.e: a certain player is having marital problems, or athlete x is using a new product that gives him an edge, etc.

How can one give proper value to that information, how can one asses the impact on such information in the final result of a game?

You apply a Bayesian prior (subjective) to a likelihood function (objective) to form a unified quantitative/qualitative Bayes' posterior distribution.

[James Marques]

You apply a Bayesian prior (subjective) to a likelihood function (objective) to form a unified quantitative/qualitative Bayes' posterior distribution.