Rankings

PoM (“Percentage of Max”) is a metric calculated by dividing the number of points a skater received for a jump or set of jumps by the maximum number of points they could have received.

One advantage of PoM is that it factors in not only GOE, but also the base value impact of any calls that a jump received. PoM also makes it easier to compare jump performance across different seasons and scale of values.

It is important to note that when PoM is calculated for a set of jumps, it is a measure of both quality and consistency. This means that a skater with memorable jumps but uneven consistency may not necessarily rank higher than a skater with less noteworthy jumps but better consistency.

Some further explanation of PoM, using a 3A as an example where relevant...

• As of the 24/25 season, the maximum score for a 3A, ignoring bonuses, was 12.00. So, as a hypothetical example, a 3A with a score of 9.00 would have a PoM of 9.00 / 12.00 = 75.0%.
• Similarly, a set of three 3As with scores of 6.00, 8.00, and 10.00 would have a PoM of (6.00 + 8.00 + 10.00) / (12.00 x 3) = 66.7%.
• When calculating PoM for a jump performed as part of a combination or sequence, the base value impact of the other jump or jumps in the jump pass is excluded from both the numerator and denominator. (It wouldn’t be excluded if you were calculating PoM for the entire jump pass, however.) For example, for a 3A+2T, the 1.30 point impact of the 2T would be excluded from both the numerator and denominator when calculating the PoM of the 3A. Also, it is a rare occurrence when analyzing ultra-cs, but if a jump pass’s GOE was based on another jump in the jump pass besides the jump being analyzed, this GOE is recalculated to be based on the jump being analyzed.
• Second-half bonuses are factored into the calculation of PoM, so to credit skaters for the increased difficulty of performing jumps later in a program.
• However, other bonuses, such as those sometimes seen in certain novice and domestic competitions, are generally excluded from the calculation of PoM.
• Repetition penalties are excluded from the calculation of PoM, since they have more to do with program requirements than jump quality.
• Fall deductions are excluded from the calculation of PoM, in part because the size of the deduction can vary based on other jumps in the program besides the jump in question.
• Popped attempts at the jump being analyzed are not factored in when calculating PoM for a set of jumps due to the trickiness of collecting data for them.
• Invalidated attempts at the jump being analyzed are not factored in when calculating PoM for a set of jumps due to there being no score associated with invalidated attempts, and due to the fact that such invalidations often have more to do with program requirements than jump quality.

Season PoM is a skater’s PoM across all of their attempts of a jump in a season.

For example, a skater’s Season 4T PoM would be their PoM across all their 4T attempts in a season. If that skater attempted five 4Ts in a season with scores of 5.50, 7.50, 9.50, 11.50, and 13.50, they would have a Season 4T PoM of (5.50 + 7.50 + 9.50 + 11.50 + 13.50) / (14.25 x 5) = 66.7%, keeping in mind that 14.25 is the maximum score for a 4T as of the 24/25 season.

Note that, for simplicity, this example assumes that none of these jumps received a second-half bonus and that none of these jumps were attempted in combination or sequence. When calculating Season PoM in practice, these are often things that need to be taken into account.

Career PoM is a skater’s PoM across all of their attempts of a jump in their career.

For example, a skater’s Career 4Lz PoM would be their PoM across all their 4Lz attempts in their career. If that skater attempted five 4Lzs in their career with scores of 7.50, 9.50, 11.50, 13.50, and 15.50, they would have a Career 4Lz PoM of (7.50 + 9.50 + 11.50 + 13.50 + 15.50) / (17.25 x 5) = 66.7%, keeping in mind that 17.25 is the maximum score for a 4Lz as of the 24/25 season.

Note that, for simplicity, this example assumes that none of these jumps received a second-half bonus, that none of the jumps were attempted in combination or sequence, and that the base value and maximum score for the jump in question was constant over the course of the skater’s career. When calculating Career PoM in practice, these are often things that need to be taken into account.

Adjusted Score for a jump is simply the “numerator” used when calculating PoM. (See the initial item above for further detail.)

In the case of a jump performed as a solo jump, Adjusted Score is generally equal to the jump score itself.

In the case of a jump performed as part of a combination or sequence, Adjusted Score is generally the score excluding the base value impact of the other jump or jumps in the jump pass. (That is, assuming that the GOE for the jump pass in question is a function of the base value of the jump being analyzed, which is generally the case when analyzing ultra-cs.)

Average Adjusted Score is simply the average Adjusted Score for a set of jumps.

Comparing and ranking jumpers across all quads presents a unique challenge because skaters aren’t being evaluated for a single jump type, but rather multiple jumps types, some of which are generally considered easier, e.g. 4Ts, and some of which are generally considered more difficult, e.g. 4Lzs.

As such, using PoM when ranking quad jumpers is not as ideal, as this would arguably favor skaters who predominantly jump easier quad types, while disadvantaging skaters who frequently jump more difficult quad types.

The advantage of using Average Adjusted Score instead is that it not only captures the quality and consistency of a skater’s quads, as does PoM, but also factors in the difficulty of the types of quads the skater performed, given that the variation in difficulty between quad types is reflected in their respective base values.

Ranking the best quads would arguably necessitate using a methodology similar to the one used for the “best quad jumpers” videos, which factors in not only jump quality, but also jump difficulty.

While this approach is perfectly sensible, the result would often be a repetitive video largely filled with 4Lzs and 4Fs.

As such, for these videos, it was decided it would be more interesting to momentarily ignore difficulty, and instead focus on the most flawlessly executed quads, regardless of quad type and difficulty.

With this in mind, the basis for the rankings in these videos is PoM.

Taken into account are all competitions covered by SkatingScores, the invaluable database of international and domestic competition results.

However, SkatingScores is missing many significant Russian domestic events. In light of this, the following competitions are also included in this data...

• All-Russian Competitions in Moscow and St. Petersburg
• All-Russian Competitions outside of Moscow and St. Petersburg (from 23/24 onward)
• Russian Novice Older Age Nationals
• Russian Novice Younger Age Older Group Nationals
• Russian Novice Younger Age Younger Group Nationals
• All Moscow Novice Championships
• All St. Petersburg Championships
• Winter Spartakiad
• Tournament for the Prizes of Santa Claus
• Moscow Open Championships, Moscow City Championships, Moscow Cup
• Moscow Battle of the Schools series
• All other Moscow events for the 1st Sports Category and older (from 23/24 onward)
• St. Petersburg Cup and St. Petersburg Federation Cup series (from 24/25 onward)
• Moscow Region Championships and select other competitions (from 24/25 onward)
• Tatarstan Championships and Kazan Championships (from 24/25 onward)
• Tatarstan Cup series (from 24/25 onward)

Finally, any competition that had one or more jumps included in the QT Club, Quad Club, and Trixel Club database is also included.

It is estimated that this data, in total, has taken into account over 95% of the ultra-cs that have been attempted in competition since the start of the IJS era.

And, from the 23/24 season onward, it is estimated that nearly 100% of ultra-c attempts have been taken into account.

For those unfamiliar, the ISU Judging System (IJS) was implement in 2004, replacing the previously used 6.0 system.

Whereas the IJS judging system assigns scores to individual elements, including jumps, the 6.0 system did not. As such, calculating metrics like PoM for jumps prior to the IJS era is not possible.

However, because my focus is primarily on the womens discipline, this is not much of an issue, since ultra-cs were rarely performed by women prior to the IJS era.

The one minor exception is for 3As, with Midori Ito and Tonya Harding having notable 3A attempts prior to the IJS era. With this in mind, both skaters have jumps included as honorable mentions in the Top Triple Axel Jumps of All Time video, with Midori Ito also included as an honorable mention in the Top Triple Axel Jumpers of All Time video.

In order to be eligible for a “best _____ jumpers” video, a skater must have a minimum number of attempts of the jump in question within the time period in question.

This minimum is always referenced in the description for the video in question, and can vary by jump, season, etc..

Setting this minimum is more of an art than a science, and often has to do with keeping the videos within a practical length.

In the case of any skater who tested positive for performance enhancing drugs, their results are invalidated for purposes of these analyses starting from the season they tested positive, and ending at the point that they were cleared to compete again by the relevant international bodies.

Note that this differs slightly from the ISU’s approach, which invalidates results starting from the point the skater tested positive, but not for prior events in the same season.

The rationale for the modified approach being used here is that a skater caught doping at one point in a season was in all likelihood doping from at least the beginning of the season in question.

For example, in the case of Kamila Valieva, there was a dramatic and abnormal statistical improvement in her quads from the outset of the 21/22 season as compared to the seasons prior. While she didn’t test positive for performance enhancing drugs until the middle of the 21/22 season, it is reasonable to believe that she was using and unfairly benefitting from the banned substance in question going back to the beginning of the 21/22 season, thus justifying the invalidation of her results starting from that entire season.