Of course balancing is important when so many cells are working together. Tesla of course knows it and they have done many things to make sure they are all stay in balance. When cells are produced they are tested and then matched. That's a very common thing when producing battery packs with multiple cells. A battery packs starts out with cells that are equal. Next thing is cooling. Cell efficiency changes with temperature. It is important that all cells have the same temperature otherwise they would drift apart. Tesla has that down very well. There are still small variations, though. For that Tesla has a balancing system. It measures and compares the groups and evens them out through resistors. The way it works is that it takes a measurement, calculates the difference in capacity and then applies the resistors to the highest groups to bleed off the amount they are higher than the others.
This happens in the background and independent of the car driving, charging or idling and is independent of charge level. I have been watching the voltage of my battery pack for more than a year now and after more than 3 years and 136k miles the average difference between the cell groups is 0.15%. That is a difference of 0.3 miles. What is more important, though, is monitoring over time how much the pack gets out of balance. What is the difference between the worst and the best days? Based on my data (well over a year checking almost every day) the difference is maybe 0.1% = 0.25 miles. In other words even on days where my pack is not balanced well it only accounts for 0.25 miles of difference. That's negligible.
One thing that is mentioned often is that balancing only happens when you fully charge. It doesn't seem to be true. According to Jason, who has taken many batteries apart and shared some of the findings here, the pack needs to be charged to aprox 93% when the balancing system takes a sample and then does it's thing. So the theory is that if you only partially charge your car for a long time and never hot 93% balancing would never kick in and the pack gets out of balance more and more. I have partially charged for many weeks, sometimes months and always monitored the cell voltages. I did not see the pack getting out of balance more. In face I was not able to see the voltage difference to become larger at all. Maybe charging to 93% (or higher) is not the only trigger point for the balancing system. Maybe Tesla improved the system over time to monitor at any level of charge. I don't know why or how, I just know the pack does not get out of balance even when charged partially for a long time.
Last but not least, there is this thought that when you charge to 100% and the charge power gets smaller and smaller towards reaching 100%. At the end it shows 99 or 100% with very little power going in and no change in charge level. People think that is balancing. It is not. That's just they way a lithium battery is charged to 100%. The charge current is reduced more and more while the voltage is kept (almost) constant. RC car battery packs are balanced at 100% by bringing up the lower cells. That's not the case with Tesla. They are using resistors to bring the highest cell groups down.
