The
most recent Perun has an interesting look at Russian tank losses by family, grouped into three-month time periods since the start of the Ukrainian invasion.
Feb-Apr 2022
T-90: 3%
T-80: 23%
T-72: 71%
Older: 3%
May-Jul 2022
T-90: 1%
T-80: 22%
T-72: 68%
Older: 9%
Aug-Oct 2022
T-90: 3%
T-80: 33%
T-72: 53%
Older: 11%
Nov-Jan 2023
T-90: 5%
T-80: 34%
T-72: 49%
Older: 13%
Feb-Apr 2023
T-90: 8%
T-80: 40%
T-72: 46%
Older: 7%
May-Aug 2023:
T-90: 5%
T-80: 53%
T-72: 36%
Older: 6%
There are a few interesting things here.
The older tanks (T-55, T-62, T-64) had a spike after the first three months of the war and have slowly declined since, probably because ones that could be refurbished quickly were thrown into combat while better tanks in deeper storage made their way to the factories for rebuilds. As the refurbishment system has worked out its early kinks, those tanks are less numerous at the front.
The T-72 and T-80 have flipped in importance. T-72 losses as a proportion of total losses have basically halved, from 71% to 36%, while T-80 losses have gone from 23% to 53%.
T-80 losses also aren't particularly new T-80s. Russia has three-ish main models, the T-80B/BV (1978/1985), T-80U (1985) and T-80BVM (2017). The BV is a B with Kontakt-1 ERA. The U has a more powerful engine, revised turret armor, Kontakt-5 ERA, and an upgraded gun and fire control. BVM has Relikt ERA and upgraded gun, fire control, and optics.
Early on (in the first three-month period) T-80 losses were 23% B/BV, 59% U, and 18% BVM. For the most recent four-month period, they were 82% B/BV, 2% U, and 17% BVM (equals 101% due to rounding). The BVM has remained pretty constant as a percent of T-80s lost even as the T-80 losses have grown, but the U has been steadily replaced by the less capable B/BV. Basically, they're upgrading as many as they can to BVM standard and then shoveling the rest of the B/BV to the front without upgrades to maintain numbers.