Personally I don't give a shit about gamblers, its a risk you take. No one other than their teammates or organization get to care.
The point I was making is that the international Dota scene pays out in USD and placing at TI is worth years and years of average income in a lot of these places. Under such a situation breaking a roster lock is an easy choice if it increases the chance to qualify at all, because the reward of qualification is so huge. Its why PED's have to be tested for in pro sports, a borderline player of any caliber is massively incentivized to cheat because if the difference between making it and not making it is PED's many many people will cheat. It was an attempt to point out why the roster lock doesn't really accomplish what they want it to, different pressures in different regions.
The main thing about China is the fact that players are owned, but also have salaries. They can be traded, but they cannot just do what they want or just be dropped without penalty/compensation. There is a legal framework, even if a bit corrupt.
Dota is a different kind of Esport, like CS:GO, Both games have had and survived engine remakes and have remained in competitive state for a very long time outside the average games competitive lifetime. It says something that these two games have remained largely the same across 10-15 years of time. Dota has seen 3 separate engines while maintaining the same game if you include Reborn.
By engine numbers Dota is becoming more popular not less. But more importantly people have played this game for 10 years and nothing implies that they will stop. Its that deep, the remakes are faithful. CS:Go the same. People also watch CS:GO and Dota who do not play it. Much the same way a person might stop playing football, but consume content their whole life. Few games have the potential for this legacy view. Most great competitive titles did not have amazing faithful remakes. For example, I would still be playing Quake in Clanmode and Threewave CTF if there was a modern update, 100% faithful to the physics that integrated into a matchmaker like Rocket League has through steam.
Basically a vehicle to capture new players is required for any "sport" to have a lifespan that continues. There must be a constant flow of newbies to allow people to learn with. Modern skill based matchmakers are amazing as long as a game has a sufficient userbase for new people to break into the learning curve. As long as this exists a competitive game will not die.
Lets look at other FPS besides CS:GO. Back in Quake, ID software propped up most of the prize pools with video card makers or a few peripheral/competitive companies like TEN/MPlayer. But as they made new versions of Quake, the old games were not supported with any infrastructure. Painkiller and other not that good but not that bad FPS had runs on game circuits because their companies put up the prize. None of these games last, because as you say as soon as the playerbase wanes and the company selling the game stops interest its over, a game will die. CS:GO and Dota have survived this time period, frankly, neither game had anyone ever propping up their merits for sales. Dota kept War 3 relevant, and Counterstrike kept Half-life relevant. But neither Blizzard nor Valve at that time put any money towards either games competitive existence.
Quake Live/3 is on Steam. Its still a reasonably good shooter, but 50 people tops are playing at a time, and no one is a scrub who will still launch it. If you have never played a day of Quake in your life, you will just lose/die and never launch it again even if you have interest. By contrast if you pick up Dota or CS this problem doesn't exist, there are enough bad players and new players to give anyone who wants a competitive match at their skill level to find a competitive match. This is key. If you went down to learn to play basketball around the corner, it would be very useless if the only person playing was Shaq. You will learn very little if Shaq just frags the fuck out of you.
The only other video game that had a chance of having this happen, was if Blizzard had faithfully engine updated Starcraft: Broodwar, it would still be played in Korea and be a big deal. However, the ancient engine stopped bringing in new players due to lack of support infrastructure, and Blizzard's desire to sell SCII put the final nail in the coffin. And lastly the single other game that could have this happen but never will is if Smash Bros was remade in modern graphics by Nintendo with all the unintentional depth and difficulty kept intact along with that matchmaker. To Smash's credit however, its following is so loyal and so dedicated that at least a few players are surviving as pros and its yearly inclusion in EVO gives it a small yearly presence. But where are the new players coming from, the giant pile of bads, it is destined to remain very Niche because only players passion keep it alive. To try and avoid people saying I didn't mention it. Streetfighter has had its own ecosystem for a long time, and Capcom's support has only gotten better. Streetfighter also has a modern matchmaker in its new iteration, new players can continue to become relevant.
League of legends has a huge following, and honestly I think it takes its own independent analysis. It has a huge player base and following, but as an Esport in the West it is a giant market tool and the players are on payroll of Riot. This is entirely different than Dota/CS. It could last a very long time, but if League at any point stops gaining/holding its playerbase it has never survived the conditions that CS:GO / DOTA survived to make it to this point.