I fully agree with you. That is why we shouldn't change our strategy only because we lost once or twice. In my opinion to fully test a strategy we need more approaches and try over a longer period. Luck plays definitely a big role in gambling that is why need to fully test our strategies before switching again. If we lose it might be because we were unlucky a few rounds. But we should also keep an eye on our winnings. It might be the case that we are just running hot and not only because of our strategy.
While there are games in which I think this makes sense when it comes to dice what kind if strategy you have in mind? There is nothing you can do to improve your chances to win, many have tried for a very long time to beat gambling games by changing their betting patterns but unless you were able to know beforehand the result of the dice then there is no point to those strategies as simulations have shown over and over again that the actual probabilities of winning or losing over an important number of rolls tends to match the theoretical values you should expect out of the game.
In my opinion the best strategy for dice games is a martingale approach. We can definitely make it work as long as we have a large enough bankroll to start with. The biggest problem I see here is that we start with too large bets initially and then are unable to cover a longer period of losses. Knowing the results beforehand isn't really possible.