What I was thinking about when reading this was that Niantic really should have made this information publicly available from the start. Seeing how you can't get more old pokemon except by waiting, it's very important to be able to prioritize their usage in trades.
Now I know that I should be sitting on the pokemon that are 1.5 years old because they're about to be become more valuable in the near future, no thanks to Niantic. Their vagueness really hurts sometimes.
GamePress
Lucky trade rates revealed!!!
The SilphRoad finally made available their study on lucky trade rates:
https://thesilphroad.com/science/chance-of-lucky-trade/
Report
Answers
Granted 9k isn't a huge sample but how many individual players have collected the data on the ages of traded pokemon for all of their trades? I have 7k something traded myself but zero of those trades are acceptable data points for a study like this.
If you look at the confidence intervals, the results for the new pokemon hold up very well. The exact rates for the older pokemon have more of a chance of being wrong, but the sample size is large enough that it would be extremely unlikely that older pokemon in the sample got a better lucky rate due to RNG.
I've gotten lucky with 12 legendary and 2 shiny. 14 special trades went lucky doing a special trade almost every day. I always trade legendary for legendary, shiny for legendary, or shiny for shiny. That seems to conform to the apprx 6% lucky rate for new pokemon. From then on, yeah I do believe older ones are more likely. However, I'm sure not going to trade something old I wasn't wanting to trade, nor save extra things to "luckify" them.
Wonder why they say the base for less than 1 year is 5%, when the lower end of their 95% confidence interval is 5.27% (and the middle 5.77%).
It may be a nicer number than 6%, and I'm willing to admit that makes it a more likely choice. However there's also the fact that the oldest 2 months are statically much more likely. So who's to say that there isn't another break at 2 months old or some other early point, and since most trades in the <1yr bucket are same day as caught those are 5% to bring down the average, while others may be 6, 7, 7.5, or 8%.
Also, would be nice if they did a test for whether the age of both Pokemon mattered or not, assuming their prior results are known. Since individually the results may not be significant, but all 3 being higher than rates with a new Pokemon would imply there's something there.
