Sure but unless you have coverage of all error cases, it’s nice to have that check at build time vs some untested corner case in production.
> Given that you have a good system for monitoring errors in production ...
Runtime errors are a totally different beast.
> ... and a low friction deploy process so you can quickly and easily fix errors when you discover them, I feel these kinds of errors are honestly not worth guarding against in most contexts outside of those where getting everything perfectly right the first time is a hard requirement (health, fintech, other safety-critical industries, etc). Of course, most companies lack either or both of these ( a good system for monitoring errors in production and a low friction deploy process), so they use static typing as a clutch to try to reduce the chance of errors making it into prod simply because they know they'd be slow to fix/detect those errors. That's a poor reason to choose static typing IMO, but I think that's by far the most common reason.
Fixing things earlier is cost effective regardless of your deploy speed. Sure it’s worth more the longer the cycle, but not having to run unit tests because the compiler knows “foo” is not “fo0” saves time as well.
This is one of the advantages of typescript over a language like clojurescript.
I'm not saying detecting errors earlier through static typing has no value of course. Just that there is a spectrum of value a team can derive from static typing that scales based on the length of their deploy and error detection cycle, and at the lower ends of that scale it's not always clear that static typing provides enough of value to justify its costs.
And that if a team doesn't already have them, investing to minimize the deployment and error detection cycle is often much more valuable than investing in static typing because it's an effective tool for both dealing with runtime errors as well as errors that could have been statically determined.
For a small team full of talented engineers (e.g. a startup) it's not unreasonable at all to just... not make typing errors. I've put 5+ large apps into production in the last decade and could probably count on one hand the number of times I've really struggled with something type related.
There's certain anti-patterns that make up the majority of type-related problems in JS. Things like inconsistently mix 'n matching falsey types, adding or removing properties from objects in unpredictable ways, or leaking stringified numbers and bools everywhere.
Other classes of type problems can be solved in dynamic languages by structuring code in more self-documenting ways and properly augmenting any remaining ambiguity with comments.
If your team is naturally coding in a way that respects the dynamic typing of the language then your benefits from TS are going to be much more marginal, and that needs to be taken into account when comparing the costs and benefits of it.
My guess is that we'll see a harder split in the JS community in the next few years. Most people are going to drift towards only taking on TS or JS roles in much the same way most JS programmers wouldn't take a lisp or C# job at the moment.
Now I avoid using regular JavaScript.
I'm impressed how many bugs it prevents, and it prevents them early.
Refactoring in TypeScript is literally magnitudes easier and quicker.
Such arrogance.. ..and unfortunately so typical for TS proponents..
It's also common to encounter errors in production that are difficult to reproduce in the development environment, this is especially true of data-type errors in dynamic languages where some unusual edge case creates a strange data structure that would simply be impossible with static types.
There's also a chance that the developer fixing the code didn't actually fix it and instead only introduced enough guard logic to handle that one specific case based on their best guess regarding the shape of data that can come through, thus redeploying with a 30% error rate instead of a 60% one, repeating the cycle of attrition testing against the production environment until the error rate is "noise" but maybe not completely fixed.
Whether or not that delta is worthwhile compared to a dynamically typed codebase with a well-optimized deployment/error detection system is not always clear however. Especially considering the costs of a static type system in limiting expressiveness compared to dynamic code and the mental overhead required when building abstractions that require layers upon layers of generics and higher order types to fully specify statically (library authors probably feel this pain the most).
I'd encourage you to keep your mind open to it's potential benefits as I've felt it has been a truly transformative experience that I'd like others to share.
Given that you have a good system for monitoring errors in production and a low friction deploy process so you can quickly and easily fix errors when you discover them, I feel these kinds of errors are honestly not worth guarding against in most contexts outside of those where getting everything perfectly right the first time is a hard requirement (health, fintech, other safety-critical industries, etc).
Of course, most companies lack either or both of these ( a good system for monitoring errors in production and a low friction deploy process), so they use static typing as a clutch to try to reduce the chance of errors making it into prod simply because they know they'd be slow to fix/detect those errors. That's a poor reason to choose static typing IMO, but I think that's by far the most common reason.