There is always a degree of limitation to such figures.
Yours rely entirely upon parents truthfully requesting holiday/time off, so automatically discounts those who lie as posters recommend to do almost daily on here. You're not differentiating between approved and unapproved absence in the case of holidays or the duration of each. It also doesn't include those who do not have KS2 data (statistically more likely for families of overseas origin).
Absences for parent reported illness include illness, undisclosed holidays, lateness where they've decided to call in sick, serious chronic illness and disability, parents covering for truancy, school refusal, waiting for EHCP, parental disengagement, domestic circumstances, lack of a place at hospital school and many, many other circumstances such as school transport failure and other issues (some of which have now been separated out into different codes for a greater depth of analysis). Illness has been used as a giant bucket for 'not coming to school - the easiest thing to say' on top of actually being ill and not feeling great/deciding it's not happening today despite the Lazarene recovery kids are famous for once it hits 9.30am and the TV goes on.
The FFT indications are an overall correlation between lower attendance = lower levels of progress in those where there is baseline data. FFT will not just base it on KS2 data, they can also now calculate it through CAT4 or other metrics, thus increasing the breadth of students this is calculated through.
As a general rule of thumb, absence of any kind results in lower progress and attainment compared to the mean, based upon thousands of students reaching data points every year. It's why hospital schools still exist (not enough, but they do still exist), because it's not the fault of a kid with Cystic Fibrosis that they are in hospital for at least 8 weeks a year for planned admissions, never mind emergency ones or a child with leukaemia isn't able to go to school; the additional education they receive there has a known, provable positive effect upon results compared to children who have not been able to access it.
Further splitting it into issues of access, transport, lack of EHCP place, train strikes, etc, etc helps to identify whether there are other correlations that could be approached differently - all the way to a local authority being able to say to a transport provider 'You are not keeping to the service levels agreed in the contract - you're not turning up for 30% of pickups, we are terminating this arrangement' and selecting one whose students have never been recorded as missing school due to the driver not turning up.
But it's all dependent upon the parents telling the truth about the one category it's easy to lie about vs specific circumstances that can be evidenced. Your claim is compromised by it not including those who lied as well as those whom for other reasons do not form part of the data set.