The Ins and Outs of TPI and Rank

The TPI formula was created by the board of directors and staff of Holstein Association USA. It is intended to be used to rank Holstein cattle and assist dairy producers in identifying superior cattle that possess a combination of high production, sound conformation and desirable health and fertility. The development of any national index is challenging. The goal of these indexes is to provide an easy way for the average farmer in the country to rank the sires (and cows) available for selection.

TPI is generally considered the industry standard, and is used in multiple countries by many producers. But have you ever sat down and considered how the index actually works, or what impact certain traits have within the index? We must realize that any index is subject to change over time – and rightfully so, as production conditions and prices change for the average farm within a country. For TPI, weightings change when HO USA determines that they more accurately reflect what breeders desire in terms of balance between all the traits. These small changes can have huge impacts on industry ranks – and therefore PERCEIVED value.

An easy way to think of TPI is to group the traits that make up the formula into three main categories with their relative weighting, as in the image above. Production traits, for example, receive the most overall emphasis (43%) in the TPI formula.

When considering the TPI index, we can say that while these weights might be the right weights for the ‘average dairy’, the real question needs to be ‘are these the right weights for MY dairy?!’

The Top TPI lists and Rank

Indexes are a very important selection tool because they combine all the important traits into one. They get us away from setting minimum criteria for particular traits, which limits genetic progress. However the WEIGHTS in the three categories that make up the index (production, health, conformation) are extremely influential in ranking the sires, making certain traits more or less important than you might perceive or value for your operation.

So what does the rank on the TPI list really mean? We know the perception is higher ranking = ‘better’, but ‘better’ is a very subjective term! Is the 100th bull really that much better than the 200th bull, for instance? Let’s look at an example.

Impact of certain traits on TPI

Assume we have 3 bulls, AltaMELHOR, Bull A, Bull B. AltaMELHOR has a sporty TPI value of 2363, ranking him in the upper echelon of bulls with semen available. Bull A, Bull B and AltaMELHOR have identical breeding values EXCEPT Bull A is exactly one point lower for each of the conformation traits than AltaMELHOR, and Bull B is exactly one point higher than AltaMELHOR for each of the conformation traits. What would you guess the TPI of Bull A and Bull B to be based on these differences in the conformation categories?

Of course to answer this question, you need to know the impact of each trait in the TPI formula. A bull gets:

Based on these differences and the information listed above, what would the GTPI be for Bulls A and B? Where would these bulls rank on the industry-wide GTPI list?

Top 100 vs. Top 200 TPI

The difference in GTPI between the top 100 and top 200 GTPI ranked bulls has become very minute. In fact, when considering the industry-wide top GTPI list of bulls >14 months old, only 39 points separate the 100th bull ranked for GTPI from the 200th bull. Additionally, only 114 points separate the 100th from the 500th ranked bull for GTPI. Yes, that means 400 bulls are within 115 TPI points of each other! This is certainly not something we were used to prior to genomics, but because so many more bulls are being tested now, it is a new reality in the industry. The conclusion is obvious; the importance of 'rank' on these lists has little significance as the difference between bulls is extremely small. 

Impact of certain traits on TPI – Example Continued

If you did the calculations, you should have discovered that Bull B would have a GTPI of 2499, and Bull A would have a GTPI of 2227. Bull B would rank #4 in the industry, and Bull A would rank #501, however it could easily be argued that the traits that have real quantifiable value (production & health) are identical across all three bulls!

Based on what you’ve learned by reading about TPI, rank, and by going through the provided example, the conclusions should be clear: 1) Conformation traits have a significant impact on the TPI formula as it is currently weighted, 2) a bull’s TPI rank will change disproportionately with small increases or decreases in individual traits because of the number of bulls now being genomic tested, and 3) the best ‘TPI formula’ is the genetic plan you develop with your Alta representative in Alta ValueBuilder. It’s easy – just put YOUR weights in the circles for production, health, and conformation! 

Article by: Nate Zwald and Lynsay Henderson