How to read the FEUD score charts on Charting the Territories

In the final week of November 2019, we introduced a second statistical metric into our analysis of wrestling in the territorial era. It's an attempt to measure feuds, ie when two wrestlers (or teams) wrestle against one another frequently over a short period of time. We have named this statistic Frequent Encounters Using Data, or FEUD for short (see what we did there?)

The FEUD score is a positive integer; the higher the number, the more "intense" the feud is at the time. Due to many factors (listed at the bottom of this post), it's hard to say what a "good" number is; as a loose rule of thumb, a FEUD score of 25 and above is a good indicator that the match in question was indeed a "feud". Many of the matches listed in the FEUD charts aren't actually "feuds"; they are merely matches that happened frequently at the time. The higher the FEUD score for a given match relative to other matches over the same time period, the bigger the feud.

Here is a sample FEUD score chart:

Just like our SPOT ratings, these are based on a weighted, rolling five-week period. So in the chart above, where it is for the week ending 12/5/82, it includes a five-week period that includes the week ending December 5th, the two weeks prior, and the two weeks afterwards. So it goes from November 15th through December 19th. The weeks are weighted differently, with the week ending 12/5 having the most weight, the weeks ending 11/28 and 12/12 having a slightly lower weighting, and the weeks ending 11/21 and 12/19 having the lowest weighting. For a match to appear on the FEUD chart, it must occur at least three times over the five week period and have a minimum FEUD score of 10.

They are listed in descending order by FEUD score. In the chart above, Kamala vs Stagger Lee is listed first, as it is the biggest "feud" in the territory at the time. The first column on the right, labeled "TW", is the FEUD score for the current week (This Week). The second column labeled "LW" is the FEUD score for the previous week (Last Week). In this case, you can see the FEUD score for Kamala vs Stagger Lee increased from a 27 last week to a 39 this week; this is a sign that the feud was becoming stronger (ie, happening on a higher percentage of cards). The third column lists the SPOT rating for that matchup; this way you can see which of these matches generally occurred higher up on the card. The SPOT rating column is also color-coded; in the chart above, there are four matches that would be considered "main event feuds" as they have a SPOT rating of .80 or above: Kamala vs Stagger Lee, Hacksaw Duggan vs Tony Atlas, Stagger Lee vs Ted DiBiase, and Matt Borne & Ted DiBiase vs Mr Olympia & Mr Wrestling II. These are all shaded the darkest. Conversely, Iron Mike Sharpe vs Marty Lunde and Kelly Kiniski vs Marty Lunde were most likely to take place lower on the cards, as they each had an average SPOT rating below .40 and thus are shaded very lightly.

The limitations/characteristics of the FEUD metric are as follows:

• Because of differences in booking philosophy (partially due the limitations of television distribution) in the 60s and early 70s, towns were often booked as stand-alone entities. That is, a main-event level match that occurred frequently in one town may not have occurred in other towns. The rule of thumb of the day was that if a match drew a good crowd in one town, they'd try and run an angle (ie non-conclusive finish) to build to a rematch the following week. So for that time period, you may see a larger number of matches with lower FEUD scores (for example, Danny Hodge or Bill Watts might be feuding with several different heels in different towns concurrently, as opposed to feuding with one heel for several weeks and then moving on to a feud with another heel).
• The "bicycling" of the TV show across the territory sometimes leads to a delayed reaction in the FEUD score. When a big angle is run on TV, it takes a few weeks for it to air in every market. So you can see a feud first appear on the chart with a low number, but over a span of several weeks it increases. Then, as it reaches its' peak and leads to blowoff stipulation matches, it might take a few weeks for the FEUD score to gradually get lower.
• There is a slight bias towards singles matches over tag team matches, particularly when at least one of the tag teams is not a "regular" team. For example, in early 1980 when The Freebirds are feuding with Junkyard Dog, JYD might be teaming with different partners in different towns. So those might not show up on the FEUD charts, or there may be two or three different matches listed with lower FEUD scores. Conversely, when The Assassins are feuding with The Kentuckians in the mid 60s, that match will likely have a much higher FEUD score.
• Much like the SPOT ratings, captains matches and cards that feature Russian-roulette style battle royals (as well as other special circumstances evaluated on a case-by-case basis) are not used. Battle royals featuring Andre the Giant or Haystack Calhoun, where they are not wrestling in an earlier match on the card and the intent is clear that their appearance in the battle royal is "the draw", are treated as a match and can appear on the FEUD chart.
• The size of the roster, the # of shows the territory ran each night of the week, the # of wrestlers on each show, and the ratio of main eventers to mid-carders to preliminary wrestlers can all have an impact on the FEUD score. When the roster is at its highest (early to mid 70s for McGuirk/Watts), there are more possible combinations of matches. This can make it more difficult for any one particular match to have a high FEUD score; the cards are shuffled on a nightly basis and two top wrestlers may not be on the same card more than a couple of times per week. Interestingly enough, when the roster is small relative to other time periods (like the McGuirk territory after the split with Watts), it is also harder for a feud to have a high FEUD score. There is often a higher turnover with a smaller roster; when they're running just one show per night, they tend to switch the matches up more often so that the cards don't seem "stale".
• The percentage of cards I have in my database relative to the total number of cards run will affect the FEUD scores. There are certain time periods where we are missing a larger percentage of cards, so those periods may see the top feuds with a lower FEUD score, because of missing data. My best estimate is that I have somewhere between 60-70% of total shows, but it varies by time period. It is possible I can add some sort of way to extrapolate and account for this in the future.
• The FEUD score can also be affected by changes in the number of shows run. For example, in the early days of the Watts/McGuirk split, it doesn't appear that McGuirk is running a full-time schedule (and if he is, we're missing a significant percentage of cards). So with only two to four shows per week, it's hard for a feud to have a high FEUD score. Seasonality is also a factor; territories generally ran a lighter schedule around Christmas. Much of the crew went home for the holidays, and they might take a week off, or run one show a night instead of the normal two or three. Again, this can make it difficult for a feud to have a high score during that time period.
• Because it is based on a rolling five-week period, feuds may appear on the chart before or after they technically began. In the example of two tag team partners that break up and begin to feud, that feud may show up on the FEUD chart a week (or in very rare occasions, two weeks) before their first match against one another. If a match that would qualify for the FEUD chart involves one or more wrestlers not in the territory (ie they finished up the week before or they start the week after), I remove those from the chart.