Tesla fans criticized cars like the BYD e6, yes. I thought it was too heavy, but I had no issue with the batteries. Others found fault with the batteries, but not those with experience with LFP like me, for instance. In fact, I did not at first understand what Tesla was doing with a clearly volatile, and fire-prone battery chemistry. Then I found out they were using special fire suppressant materials in their packs, and they used fusible links in series with each cell so that if a cell shorted out, it would blow the link. They also worked on their cell caps to reduce pressure in the event of outgassing. None of that could stop the greater volatility and susceptibility to thermal runaway and pack fire. It only reduced the spread speed, and gave time for occupants to exit. They armored the pack against puncture with a thick aluminum pan and gave it strong side supports. Every NMC and NCA pack will catch fire if punctured. Winston showed movies of LFP punctured by nails and bullets. Those are standard tests. There is a flame test, a bake test, and a full short test. (Don’t try this at home.)
Done right, LFP can be punctured with no fire. Tesla just went ahead and used NMC and NCA anyway, and dealt with it with clever countermeasures. As NMC went to more energy density with higher nickel content, it became even more volatile.
On top of the safety benefits, LFP has much higher cycle life, a minimum of 4,000 cycles, a stiff discharge profile, and very low internal resistance. It is a really good power source. Those of you with LFP Teslas or other vehicles, hang on to them. Their battery packs could last a lifetime.
Something to consider is that NMC and NCA cells can use less volatile electrolyte, with several solutions to that particular problem already existing in commercial applications. On top pf that, there are additives to electrode material that can physically separate them from their current collector, thus stopping all electrical flow completely. Panasonic is starting to produce such cells now.
LFP doesn't have a higher cycle life, it has an equal cycle life but with lower current capacity. So at the end of the cell's expected lifetime the LFP cell has stored and discharged vastly less Watt-hours than an NCA or NMC cell. This is a really critical component of the lifetime cost calculation, obviously.
LFP isn't a really good power source, it's an entirely acceptable power source. But in the application of a vehicle where weight is a key factor because road wear increases with the 3rd or 4th power of vehicle weight, any cell technology that requires a physically larger and heavier pack (LFP requires both dimensions be increased) isn't a clear win. LFP is a great technology for stationary storage, though, since the size and weight are much less of a concern.
One of the key things to remember is that any technology that improves one cell type is likely to improve all similar cell types. So as improvements in electrolyte and current collector safety improve with LFP, they also improve with NCA and NMC which makes the latter two more attractive.
In the recreational marine market, where fire is a beyond catastrophic event, everyone seems to have standardized on LFP for its safety characteristics. We're replacing lead-acid (mostly AGM or sealed deep-discharge variants) batteries anyway, so just going to anything lithium-based is already a huge improvement in terms of weight, volume and storage performance.
LFP in Marine applications is mostly a cost driven choice, because the cost per kWh is 10-20% lower for LFP compared to NCA and NMC. NCA and NMC batteries are absolutely available for marine applications, and they store around 33% more energy for the same volume as LFP which makes them a good choice in space constrained applications. But, as you pointed out most marine applications are converting from AGM which is so huge that any Lithium based battery is a win.
If fire safety was the determining factor, LTO would be the battery choice rather than LFP. But nobody's going to make that choice because LTO sucks compared to LFP.