These scores are meant to give you some idea, at a glance, of whether a particular line of TPS might be right for you. These are based on actually growing out the seed and evaluating the progeny, so some lines of TPS lack scores because I haven’t grown them in an amount sufficient for analysis yet. Most TPS packets contain 100 seeds, so this is, more or less, the number of seeds per packet that will give the desired result. Some mixes have a reproducibility score and others don’t, simply because some mixes don’t promise any particular trait.
There is nothing intrinsically superior in a high score. Generally speaking, a line of TPS with a low reproducibility score will have greater diversity, which might be something you want. On the other hand, if you are determined to get a potato just like the parent clone, then a high reproducibility score will be more attractive to you. A line with a high practicality score is definitely going to produce more food than one with a low score, but lower practicality is often the price of working with more unusual traits. So, if you are focused on selecting really unique potatoes, you are probably willing to work harder to find the varieties that combine those unusual traits with reasonable practicality.
Percentage of progeny that will have a phenotype substantially similar to the parent (or promised trait in something like a red potato mix, in which case this is the number of potatoes that you can expect to be red). Similarity is subjective, but if the parent is high yielding, starchy textured, with blue skin and flesh, I would consider any progeny that are high yielding, starchy textured, with blue skin and flesh to be similar. They will, of course, differ in many ways, such as shape, maturity, flavor, disease resistance, etc., but the general idea here is that you will get something that looks similar to the picture of the variety and has the traits that I describe as being characteristic of the variety. Self-pollinating tetraploids tend to have high reproducibility scores, while self-incompatible tetraploids have lower scores, and diploids and wild species hybrids tend to have very low scores.
The percentage of progeny that will produce more than a pound of tubers when grown in the second year from tubers and that have at least three months of dormancy. If you are not doing further breeding and you are just interested in selecting some varieties from seed to keep growing, this score tells you how easy the job is going to be. There are not a lot of surprises here. Tetraploids from modern or Chilean parents tend to have high practicality scores, while diploids, Andean tetraploids, and wild crosses tend to have low practicality scores.