I think that there are two problems with this:
1. You are not going to find a lot of profitable arbitrage opportunities. Even if you sign up to a subscription service, by the time that you see a recommendation/tip it's probably already too late and the opportunity has been taken advantage of.
2. A lot of the people here don't have the bankroll management that is necessary for this, and it's not a "surebet" in the sense that you are only making paper profits (in terms of EV) and not necessarily booking it in reality. There is a huge difference in risk.
1. The theoretical possibility is one thing, the practical implementation is another. I showed a theoretical possibility. And in fact, arbitrage opportunities are not fleeting phenomena at all - for example, if you bet a week before the event, then you will most likely have the opportunity (long in time) to make an arbitrage counter-bet. If this opportunity does not appear, then you can wait for the start of a sporting event and catch such a bet in live mode.
2. The problem of the presence or absence of the necessary bankroll does not cancel the fact that there is an opportunity to make arbitrage bets.
I do not see a fundamental difference in risk, since at a distance (as I showed in the calculations in the first post), from the point of view of mathematics, such bets are equivalent to classical arbitrage.
You're right for 1. Theoretically, there is nothing different when it comes to EV and yes I do agree that some arbitrage opportunities can be long-lasting, but these are definitely rarer to come by.
I disagree with your second point however.
There is an inherent difference between a surebet (which returns you a profit, without risk) and what you've shown. Yes, EV is positive but you have significant variance in your outcomes. You always have a non-zero risk of losing all your bankroll with this.