In a recent episode of their show Summoning Insight, MonteCristo and Thooorin postulated on shotcalling and posed the question, "Why is it so hard to find a good shotcaller?" To answer succinctly, it's because taking up the role of the player who calls the shots makes you play worse. Our brains have a limited amount of bandwidth they can access at any given time. Whether it's conscious, subconscious, or even unconsciously accessing information, there is only so much that your brain can do at once.
Oftentimes in professional sports, you may hear a commentator or coach speak of a player "Thinking, and not reacting" when they appear to be playing worse. And that sentence alone speaks volumes about why shotcalling in competitive gaming is such a difficult task.
I mention David Sirlin often in my writing, and I'm pretty sure I've mentioned this particular article several times before, but...Sirlin wrote a blog about Intuition, the Mental Iceberg, and Narrow bandwidth of players when it comes to balancing multiplayer games. The points made in his article can all be applied to players as they play the game to account for the difficulty of shotcalling.
To summarize, the Mental Iceberg is an idea that players have the amount of game knowledge they can explain (the tip of the iceberg), and then a massive unquantifiable amount of game knowledge they can only know instinctively, and not really describe or explain (the underwater portion of the iceberg). The underwater portion grows constantly with experience, while the tip's size may stay the same; a player who can clearly explain why they do everything they do may not have a very deep amount of knowledge under the surface, they are just better at verbalizing the knowledge they do have.
When you do finally try to put the reasoning or logic into words, you're narrowing your manner of thinking. Despite the logic required to make a decision being based off subconsciously assessing many factors, when you try to describe it, you're almost guaranteed to focus on a smaller amount of variables than you actually considered.
Have you ever been asked "Who sings this?" and, despite KNOWING who the artist was, were unable to answer? The answer escapes your brain and hides from the seeking monster of conscious knowledge. As you attempt to access certain information that is stored in your brain, it can hinder the actual thinking process.
Back to the point, a player who is tasked with calling the shots for a game, unless they are operating and communicating purely out of instinct or habit, are spending their entire game thinking. They're accessing the tip of the iceberg, while the underbelly remains unscratched.
Beyond this limiting the game knowledge they actually have, it also hinders their mechanics as well. Except in the case of a true genius player, you will probably find a staunch correlation between the player acting as a team's shotcaller and that same player performing less impressively than his team when viewing games at a professional level.
There is evidence of the same happening in console shooters -- players like Walshy, GhostAyame, or TeePee generally didn't put up fantastic stats in games. But if you asked a player on their team why they won, they would often give credit to those players. When I played Halo, I competed in Big Team Battle (8v8) tournaments, and would often act as the shotcaller for my team, determining plays such as when to take a Warthog, when to push certain areas of the map, where to take the flag. My team would, on occasion, upset better-skilled teams (early on, Master Theory was not the most mechnically talented of teams) due to our superior map movement and strategy. But at the end of the game, I would be lucky to have above a 1.0 Kill/Death ratio. I was a player who constantly tried to ensure the other 7 players on my team were on the same page as me and my strategy, and my own play suffered for the sake of a win.
I think Cloud 9's idea of breaking down the shot-calling into multiple aspects is borderline brilliant, since it limits the burden on any single player and frees them up to each play individually better.
The anime Log Horizon mentions similar battle roles apart from a player's actual character role they are in charge of tracking in battle. Apart from a party's formation being responsible for ensuring monsters attack the tank instead of the damage dealers, two major parts of a party's combat ability fell on the Operator and Field Monitor. The operator's job is to monitor the status of each combatant, keeping track of their ability to keep fighting. The field monitor's job is to monitor the area, tracking enemy movements and reporting them to the party.
Cloud 9, to some degree, splits these two roles between Hai and Meteos. Meteos takes charge of shotcalls regarding buffs, dragons, and barons, timing when they respawn and deciding if they team should contest. Hai controls player lane assignments and rotations, while also being the loudest voice in a team fight, calling who to focus.
The biggest problem that arises with good shotcalling is that...when you encounter a team that is either mechanically superior to your own team or has superior shotcalling, and you begin to lose to them, your team's shotcaller will always be the first ignorant fans point the finger at for the chopping block. Statistically and mechanically, the shotcaller will almost always look to be the worst player on the team. (If this is not the case, then you have a shitty player on your team that you need to replace anyway.) So the role of shotcaller is a stressful one. You'll be constantly under scrutiny if your team is not the team on top of the scene. If Cloud 9 starts to struggle, Hai and LemonNation will likely be under the Reddit microscope. If CLG slips, Dexter will have to bear the brunt of angry fans.
Reginald was, for the longest time, the primary shotcaller of TSM. And for the most part, he was a pretty damned good shotcaller. Mechanically-speaking, he flopped a lot. He was good enough to compete with the North American scene for the most part, and even held his own at S3 Worlds against Faker in lane. But he wasn't really good enough to single-handedly take over games. But he was also shotcalling. When he replaced himself with Bjergsen, TSM didn't really gain much ground over other teams in the region, and they've been criticized quite a bit for their shotcalling in the later stages of this season. Whenever TSM performed poorly, the blame often fell on Regi (except for Worlds where a lot of fingers got pointed at Dyrus for shitting the bed top lane). He's mocked for his Blue Cards on Twisted Fate, but his engage calls by pulling a Misaya on TF were a large part of his 100% win rate as the Card Master.
So my emphatic explanation of why shotcallers are hard to find in League of Legends (and basically every other game you'll encounter) is that when you find someone who's actually good at calling shots, they are always the players who look weakest mechanically. When a team's success rate falls off, they are always the first on the chopping block BECAUSE they look weak mechanically. It is extremely difficult to find a player who can shotcall efficiently at the highest level without being an exploitable weakest link on a team.