Faulty Bottle Rates

This one's for the intersection of wine and number geeks and goes outside of my typical Chicago focus.

Inspired by a thread on the Wineberserkers forum, I took a pass at rates of wine bottles with a fault to them.  Many factors could potentially cause a bottle to show poorly--winemaking issues, heat in shipping or storage, cork taint, and many others.  I pulled data from CellarTracker, where users can flag spoiled bottles, to build the tables & charts that follow.

First, I started with a look at whether some regions of the globe are more prone to these issues.  The summary below shows the "Flawed/Defective" rates across Regions.  Given the myriad ways a bottle could be flawed, 1.3% overall doesn't seem too bad, and there is fairly minor variation from that average.




My next step was to look at White Burgundy, where some of the world's best Chardonnay is produced.  Unfortunately, it also has broad notoriety for Premature Oxidation ("premox", "pox"), an issue in which wines advance in age much more quickly than expected.  This can be quite disappointing when opening a bottle after many years of holding in the cellar only to find a flat, unappealing liquid.

Isolating white wine in Burgundy (~100,000 notes in CellarTracker), the flawed rate is 3.42%, two and a half times the norm.  Worth keeping in mind: 2010 White Burgs in CellarTracker have a median price of ~$55.  Premox issues first gained notice with vintages in the mid-nineties; this rise is clearly shown below.




Continuing to look at % of flawed bottles, here looking at % of bottles opened at a given age...we can also see in the colorful chart below that, since the mid-nineties, off bottles become increasingly common past the 5 year mark.  That's not especially great for wines that are frequently hoped to be long-lived and are rarely found in the bargain bin.  Additionally concerning is that, while there is some ebb & flow to the numbers, the problem is not getting convincingly better.
Ex.: Look in the "7" column and you'll see that the 05-07 vintages are not less likely to be flawed than 99-01 were at a similar point of their development.




Lastly, I examined it across producers (admittedly imperfect; my text mining isn't as polished as I would like), finding hundreds of producers--albeit often with limited sample--producing off bottles at rates many multiples above the average.

All data came from CellarTracker, which is a fantastic site, but we should keep in mind that this is crowd-sourced and there may be varying influences on the "Defective" flag--personal experience, style preference, storage conditions, etc.  It's also possible that multiple notes represent a single bottle (good or bad).  That said, there's enough here for me to mindfully avoid some producers and to consume my White Burgundy bottles mostly by the 5 year mark, and certainly before they reach age 10.

Edit: 

Based on feedback from posting some of this information on WineBerserkers,  I took a look at running/cumulative flaw rates for a vintage.  Essentially, once a vintage starts to get in the orange/red zone of the table, things are getting more risky.  Read this left to right, and see cumulative % of White Burgundy bottles from that vintage that have been flagged as flawed.

And in this line chart I took a cross-section at several points of a white Burg's life., if the lines slope downward, the issue of premox is improving.  Here, while they are sloping down, it is a rather minor slope & seems too early to say that producers have progressed in solving the issue.