if server time is out of sync - that introduces interesting shift in price action

(bit weirdly as exchange is supposed to reject out of sync requests or something)

doji candles on backtest becomes a part of previous / next candles on actual performance graph

whole thing becomes a mess and I think is a culprit for the drawdown algo experienced

changed `price()` to `MedPrice()`, backtest results unchanged, synced time

still not a perfect 1:1 to backtests, but at least now I see an actual resemblance

-------

trading business ain't doing well lately