Apologies in advance for the rough and ready nature of this post.

In the comments on TPalmer's wonderful Raul Meireles Statistical Breakdown, someone decided that due to the relative ball retention percentages and some other convoluted logic that, even when considering the differences in price, Chelsea should have bid more money for Modric because "his market value if he was put up for sale would be about £60m".

I have taken the liberty of collecting the data on 'ball-retention percentage' (BR%) of a few midfielders transferred this summer and it turns out that there's a pretty good correlation between BR% and transfer fee.

Here's the data that I've found for transfer fees and BR%:

Charlie Adam - £7m - 46.5
Charles N'Zogbia - £9.5m - 44.6
Mikel Arteta - £10m - 70.6
Raul Meireles - £12m - 53.1
Ashley Young - £15m - 49.8
Stewart Downing - £20m - 53.3
Samir Nasri - £25m - 61.0
Cesc Fabregas - £26m - 66.9

These fees are the 'reported' values; some of them will be estimates if the official fee has not been disclosed, but they're the best numbers we have. Anyway, when these are put into a chart, it's easy to see the relationship between the two variables:

It's not the greatest of charts (just knocked it out in Excel), but excluding the obvious outlier, Mikel Arteta (whose transfer fee was lower than his skill would suggest because of his injury history and how desperate Everton are for money) there is a reasonable correlation here between BR% and transfer fee.

It is therefore not unreasonable to conclude that based on Luka Modric's BR% of 62, a reasonable 'market value' would be much closer to £25m than the £60m previously claimed.