Sometimes I use blitz as a training tool, for "training" a "live" position. For example today in this game :
https://en.lichess.org/Wq5Zrtg9/black#63
I was wondering : If I give back the exchange, will the pawn ending be won ? At first I was a little bit in doubt. But the black king was in better shape than the white king in the pawn ending, so ... I decided to just test it.
And it turned out fine, as white got into zugzwang and had to give up a pawn, resulting in a clearly lost pawn ending.
I think this testing in "live" games can be useful, because the impact of playing a line live in a game is bigger than when analyzing just a bit afterwards, and I think it can also help to improve intuition as well as the courage to make certain decisions (pawn endings can be very very tricky. one mistake can mean a loss instead of a win).
Apart from these reasons, I think such an approach can also make your own chess more interesting.
Only playing games to win a few rating points and not really looking at the content of your games is different than consciously working on your chess improvement and also enjoying the chess itself.
For example "Ha, great, I won against 1500, flagged in lost position" sounds different compared to e.g. "Hey... This position in my game was new and interesting and reminded me of the chess study by Afek I saw last week". :)
h f ! \o/
