I have, at best, a 50% success rate with Flight Simulator's Live Weather function. On two occasions I have seen FS inject what is very clearly 24-hour-old met conditions (or hopelessly wrong current data). 

So I pay for Rex Weather Force. Add a few airports to the Favorites list and go exploring. Two airports have nil data (and claims to have last been updated on Monday 1 January 0001 😏). One of those is a METAR station (and yes, the METAR is being broadcast to other sites). Other airports list state the last update as 26 hours ago. 

How is this good enough?

How do you know the MSFS live weather data is wrong? What is your benchmark? Are you using your own eys out of your window for instance?

CJ

Yes, comparing against what I can physically see (and read in terms of wind direction), along with published METAR. 

Metars are dated and not on a single coordinated update schedule.  The automated ones are universally poor.  I find this often in the real world flying I do.  I have seen it listing 400' overcast on clear blue skies only because the adjacent farm land was doing a controlled field burn off (common in my area).  Understanding this has made me think that the current 'modeled' approach from Meto (both for this sim and their aviation payware package) are the better way to go.  On their own site they discuss how Metars are only a 'component' of their overall analysis.  However, I think there is a time lag issue probably associated with the info path Meteo-MS-You.  It would be cool if the in game info came directly from the Meteo servers, but it does not. 

Just take a look at SkyVector.com and you will often see a larger airport with 'overserved' conditions showing a normal VFR, yet just a mile or two away an 'automated' shows IFR.  This variation is very common and one has to take the automated ones with a 'grain of  salt' so to speak.

I'm not sure if this always the case but I've had better luck flying in Europe, than in N America.