Taxi Driver Online

I wish they had asked someone with a brain to go through the results to weed out the rubbish, but maybe they didn't know anyone with a brain.

_________________
IDFIMH

Did a bit of analysis on the 'Regional fuel types' table earlier, but a slight technical failure meant I lost it all

Might have another go tomorrow, but can't be bothered just now. But just had a quick look at the wheelchair accessibility table...and, OMG, absolutely no commentary required

No prizes for guessing which column is most doolally

But good to know that 60.00% of London dual licences are wheelchair accessible. But what precisely is a dual licence in London anyway?

But, I mean, nearly 58% of PHVs in East of England are wheelchair accessible, but only around 4% of Yorkshire and the Humber PHVs are wheelchair accessible?

And shouldn't the figure for London HCs be 100% rather than 83.3%?

Just noticed that in fact the survey did say this about Ola:


Which begs the question, why did they include it in the results? (Survey was conducted in July 2024.)

And pretty sure Ola never actually launched in the two areas in which it was supposedly most popular (the South East of England and Scotland).

And I don't believe any of the other most popular apps for drivers in the regions either (except where it's Uber).

I mean, the most popular app amongst drivers in the West Midlands (ie Birmingham, Coventry, Wolvherampton, Dudley, Walsall etc.), for example, is something called TaxiApp UK? Seriously?

The DfT stats actually show specific numbers of WAV PHVs and of course total PHVs - I'd guess that's maybe something to do with the records councils have to keep regarding the WAV stuff and the Equality Act?

So I did a wee analysis comparing the Dft WAV PH numbers as a %age of total PHVs (the DfT stats don't show the %ages), and compared them to the new survey stats for the two highest and lowest regions in terms of the proportion of PH WAVs.

The DfT stats I'd guess are consistent with what we intuitively think would be the proportion of WAV PHVs.

So where the new survey gets these figures from is anyone's guess. But, unless I'm missing something spectacularly obvious, the new stats are just nonsense. They're a bit more credible as regards the bottom two - the South East, and Yorkshire and the Humber. Of course, there may be differences in how they compile the figures in terms of definitions etc. But, as regards the other different regions as a whole, they're not even in the same ball park

(Drafted this yesterday and there may be one or two errors, but I'm sure anyone reading will get the general idea )



So this table is across both codes. And although I don't get out of my area much, this looks immediately problematic. I mean, around 1 in 4 of all cars in the Scottish industry are 'fully electric'? My ar$e

Or, for example, more than 1 in 3 of all cars in Northern Ireland are 'fully electric'? Had a look round some official stuff and despatch office websites via Google, and found zero evidence of that. Zero evidence to support claims regarding zero emissions, sort of thing

The DfT does in fact produce spreadsheets about this kind of thing, and usefully seems to use the same regions for the analysis. Problem is, the Dft produces separate figures for both codes, and also uses different and more detailed categorisations, so difficult to compare precisely with the figures above.

But take the East of England figures above, for example. Supposedly 20% of cars between the two codes are fully electric. So 1 in 5 HCs/PHVs are 'fully electric' in the area covering the likes of Essex, Cambridgeshire, Suffolk, Norfolk and Hertfordshire

According to the DfT stats, around 80% of HCs in that area are either petrol or diesel, and around 70% of PHVs are either petrol or diesel. Yet according to the table above, across the two codes only around 41% are 'fossil fuel'.

Make it make sense

The DfT has a category called 'battery electric', which I'd guess should be the same as 'fully electric' in the table above. Yet the DfT stats for 'battery electric' are tiny in all regions, and across both codes - HC and PHV. Thus all regions are about 1% or so 'battery electric', or maybe 2% maximum. Intuitively, I'd guess that's what most of us would think from personal experience, and the 'full electric' stats from the table at the top of this post just don't reflect reality.

And, generally speaking, in almost all regions the DfT stats show both codes as being at least 60% petrol or diesel, and up to 80% or so for HCs.

Compare that to the 'fossil fuel' numbers in the table above, and they just can't be reconciled. Generally speaking, apart from simply petrol or diesel, the DfT's stats show very roughly 30% of cars across both codes as being some sort of hybrid or mentions the word 'battery' as regards fuel. Which is just way out of kilter with the numbers in the table above.

(Worth pointing out that in the DfT stats London is definitely an outlier in the context of the figures as a whole. This is because what they call 'range extended electric' make up 53.7% of London HCs, while the figures everywhere else are tiny - 1.6% at most, down to 0% in the North East - but that's simply all about the LEVCs, which aren't fully electric vehicles.

By the same token, 14% of London PHVs are 'plug-in hybrid electric', while the numbers for this everywhere else are tiny. Not sure precisely what that's all about, but obviously a quirk for the capital there. I mean, that's presumably about cars that are plugged in, but still primarily either petrol or diesel powered. Not sure precisely why there so many of those in London, but very few elsewhere.)

This is the link to the DfT spreadsheet I was looking at. The stats are to 31 March 2024, so just three months or so before the TaxiPlus/TaxiPoint survey, thus the gap seems highly unlikely to be, er, statistically significant in terms of explaining the variances.

The numbers are shown per region, rather than each local authority. But they seem to use the same regions as TaxiPlus/TaxiPoint, so presumably they should be at least roughly comparable.

Only caveat is that the DfT spreadsheet seems to be saying that the numbers are organised by region via the postcode on the VR5/registered keeper, which might not concur with licensing authority areas, since the VR5 will generally be the driver's address rather than where it's licensed. But the details of how the numbers are compiled aren't clear at all, so I could be wrong about that. But, in any case, the two surveys are still in two ball parks so obviously far apart that they can't be reconciled by minor deficiencies in the DfT stats, I'd guess.

Incidentally, the DfT spreadsheet below is typically difficult to navigate - you need to go to the tabs towards the bottom right to see the actual stats, and the actual tab might not be available without a couple of clicks. And there's a separate tab for both 'taxis' and 'private hire', with both usefully containing the letters TAXI in the tab description

https://assets.publishing.service.gov.u ... xi0115.ods

Genuinely I'm impressed by the amount of time SW is spending on dissecting this survey.

I take the view that it is an utter pile of sh**, which might miff those behind the survey, but come on they should be embarrassed by it.

_________________
IDFIMH

Thanks - and I've only looked at about four pages yet

Seriously, though, not going to spend too much more time on it, because that could take weeks, and in the final analysis, all pretty pointless.

Of course, there's nothing to suggest that fundamentally any of the numbers are actually incorrect. But they just don't seem representative of the trade as a whole, and surely the point of it all was to produce a representative survey

So maybe the sample was fundamentally skewed in some way, or too many of the respondents misinterpreted the questions, or maybe whoever did the table etc misinterpreted the answers. Would be instructive perhaps to look at the questions again and maybe work out how people completing the survey could have misconstrued the questions and thus provided misleading data...

Another outlier with the London numbers that I should have mentioned earlier is that 20% of PHVs are 'battery electric', which is way above the average elsewhere - the DfT says 1.2% of all PHVs outside London are 'battery electric', which sounds realistic.

But this means that the DfT's summary stats show that 8.8% of PHVs in all of England are 'battery electric', which is a bit more consistent with the new survey's numbers in the table further up this thread. But this is largely skewed by the London quirk, and elsewhere in England and Wales it's only 1-2% or so, thus completely out of kilter with the new survey's numbers.

Also saw this in the DfT's stuff: "battery electric: this includes fully electric vehicles only".

So what they call 'battery electric' is in fact the same as the 'full electric' column in the new survey, thus confirming that the definitions for that category are the same. And, to that extent, assuming the DfT figures are in the right ball park, the TaxiPlus/TaxiPoint stats are off the park

There might actually be an element of truth in that number, especially if that number includes hybrids.

In fact it might even be an underestimate.

_________________
IDFIMH

Sussex, the DfT's stats for London PHVs deffo show the 'pure' EVs at 20%, and nothing to do with hybrids, and as should be obvious from the extract below it's an outlier as regards England as a whole.

Ditto the 'plug-in hybrid electrics', which are pretty well negligible elsewhere. But, as a consequence of those two categories, petrol and diesel PHVs in London are very low compared to the rest of England

(The numbers are %ages, obviously, and I've left out the 'others' column in case anyone's wondering why the numbers don't add up to 100%, but in most areas the 'others' column is negligible.)


PHVs in England @ 31 March 2024, extract from DfT stats




These are the DfT's HC figures below, and the London HCs are also an outlier, for obvious reasons - the 'range extended electric' are presumably the LEVCs, and pretty well negligible elsewhere. And zero petrol HCs in London, for obvious reasons.

Not sure what the 1.3% 'battery electric' HCs in London are, though


HCs in England @ 31 March 2024, extract from DfT stats

...anyway, I've done a bit of number crunching with the DfT's stats to compare them to the TPTP survey (TaxiPlusTaxiPoint ) and tried to reconcile them.

Can't be bothered tonight, but will put them up tomorrow or over the weekend. But for the reasons above, I've missed out London from the comparisons, because London's deffo a bit of an oddball compared to the rest of the country. Won't include Scotland either, because the DfT stats don't include Scotland, and I don't think there are national figures of the type produced by DfT.

I'm guessing there will probably be around the same amount of people reading this thread as will delve into the nitty gritty of the DfT's various stats and spreadsheets. Thus very few people indeed

But here's my attempt to reconcile the new TP stats (TP = Taxi Plus or Taxi Point - take your pick) to the DfT's numbers for 'fuel class'. Which is obviously quite an important topic these days. And more difficult to provide figures than when cars in the trade were either petrol or diesel, and no real ambiguity between the two.

And it's maybe more a case of 'comparing' them rather than 'reconciling' them, because the TP and DfT stats are in completely different ball parks, as a quick glance at the TP tables above will demonstrate to the vast majority within the trade, even without looking at the DfT stats

First issue is that the DfT stats are as at 31 March 2024. I think TP's survey launched in mid-July and ran for ten weeks, but presumably most of the responses would have arrived early in the survey period.

Anyway, there's still a gap of several months between the two, and there's not a lot that can be done about that. Of course, the general trend in the trade is away from fossil fuels and towards hybrids and EVs etc, so to the extent that the TP survey numbers show more of the latter than the DfT's, then the few months' gap between the two might explain the difference. But only 'part' of the difference, surely, because the basic numbers are too far apart to be explained by a couple of months between the two sets of figures.

However, there are a couple of more important reasons it's difficult to compare the two.

First, TP uses three categories for 'fuel class', while the DfT uses seven

However, I think it's probably straightforward enough just to add the DfT's categories together and then compare them to TP's categories and numbers. But, for the avoidance of doubt, this is how I've compared the categories between the two surveys.

And here again are the graphics showing the TP categories, and the DfT categories (here for just HCs):






TP - 'Fossil fuel'
DfT - 'Petrol' and 'diesel' - not a lot of room for doubt there, surely?

TP - 'Hybrid'
DfT - Here I've added together the DfT's 'hybrid electric', 'plug-in hybrid electric' and 'range extended electric'. The latter seem to be the LEVCs mainly in London, and at a rough guess I'd have assumed they'd be the same as 'plug-in hybrid electric'. But clearly not, at least as far as the DfT are concerned. But they're surely not what TP calls 'full electric' either, so presumably TP would include them in their hybrid category. (But because London is something of an odball - as explained earlier - they're not included in the analysis below. And because 'range extended electric' are negligible outside of London, there's not really any potentially statistically significant issue with just adding them into the general 'hybrid' category for comparison purposes.) The DfT also has another column labelled 'other' not shown in the extract above, and that may be the gas, hydrogen fuels and suchlike. But the numbers are tiny, so I've just added them to the DfT's other hybrid and similar categories.

TP - 'Fully electric'
DfT - 'Battery electric'
I'm assuming those two are simply the same thing, and thus like the fossil fuels category it's pretty unambiguous. And elsewhere in the DfT's stuff they say: "battery electric: this includes fully electric vehicles only". So I think it's fair to assume that those figures are comparing the same thing in both the TP and DfT - what we'd maybe just call EVs.

So basically TP's categories of 'Fossil fuels' and 'Fully electric' are pretty easy to compare to the DfT stats. All the other DfT categories I've added together to compare to TP's 'Hybrid' column above. There may be some oddballs like hydrogen and gas, but the numbers don't really look significant (accept for maybe the West Midlands, which shows 2.6%/2.7% in the DfT's 'other' column for HCs and PHVs respectively. Could they be gas powered? Yorkshire and Humber also feature in that same category as over 1% for both HCs and PHVs, but apart from that the figures from all the other regions in that 'other' category are miniscule.

So, essentially, the first major issue when comparing TP to DfT is how they categorise the 'fuel type'. The second major problem is that the TP table above doesn't differentiate HC and PHV - it's presumably just an amalgam of the two codes in each region.

On the other hand, the DfT provides separate stats for HC and PH. And it's not simply a case of taking an average of the two - for example, if in region A HCs are 70% fossil fueled and PHVs 80% fossil fueled, you can't just say that for region A the average is 75% fossil fueled between the two codes. The reason why you can't do that should be obvious from the table below, which I've compiled using the DfT summary stats:



So the problem with simply averaging the two percentage figures for HCs and PHVs should be obvious - in each region HCs are always outnumbered by PHVs, and thus simply averaging the two figures would be misleading. This should be most obvious in the case of the obvious imbalance in the West Midlands - if, for example, HCs there were 80% fossil fueled, but PHVs were just 40% fossil fueled, it would be misleading to say that over the two codes the average was 60%, because the vast majority of cars are PHVs

So instead a weighted average should be used to reflect the ratio of HC to PHVs. So suppose HCs are 80% fossil fueled and comprise 10% of the combined fleet. And PHVs are 40% fossil fueled and comprise 90% of the combined fleet. The average fossil fueled over the two codes isn't 60%, and intuitively it looks like it might be a bit less than 45% or so. In actual fact, crunching the numbers provides the weighted average of bang on 44% - ((90% x 40%) + (10% x 80%))

But, basically, that's how the figures that will follow have been collated from the DfT's stats to compare them to the TP survey, and I'll post the actual figures over the weekend.

And, by the way, no prizes for guessing why the HCs in the West Midlands comprise only 8% of the combined licensed fleet - it's not so much that HC numbers are especially low, but because one particular council in that area licenses a huge shedload of PHVs

In fact, the HC proportion would probably be quite low anyway, because the West Midlands is a very urban/metropolis type of area, where HC numbers tend to be quite tightly controlled, and the market is served largely by PHVs (Birmingham, Coventry, Wolverhampton, Dudley, Solihull, Walsall, Sandwell, Stoke...), but you-know-where is certainly a further distorting factor.

Note that the next lowest ratio at 14% HCs is Yorkshire and the Humber - it's got similarly urban cities and metropolis-type areas like Leeds, Bradford, Hull, Sheffield, Wakefield, Kirklees (Huddersfield, mainly), Doncaster, Rotherham, Barnsley, Calderdale (Halifax, mainly), blah, blah.

(In fact if you drill down further, the proportion of HCs in some of those places is tiny - 4% or so in Bradford. Indeed it's likely to be even lower than that in reality in view of presumably a significant number of PHVs operating there but licensed you-know-where )

The only other region with HCs below 20% of the total fleet is the North West - that's Merseyside, Greater Manc, Wigan, Warrington, Blackburn, blah, blah.

So after all that rigmarole, here are the comparison tables - since TP (TaxiPoint/TaxiPlus) have produced three different fuel category stats, there are three comparison tables, because otherwise it all becomes a bit cluttered. But this is the first one, and is the simple comparison of TP's 'fossil fuels' figure with the DfT's combined petrol and diesel figures. No room for ambiguity in the definitions there, right?

And remember that the London, Northern Ireland and Scotland figures provided by TP aren't included below.

And I haven't bothered with a title or dates, etc, because it's just a rough comparison. But, roughly speaking, all the figures are from mid-2024.

So basically it's the two columns in bold that need to be compared. The first two columns are the DfT stats separately for HCs and PHVs. The third column is the weighted average of the DfT's stats between the two codes



Anyway, the most obvious takeaway from the above is that the TP figures grossly understate the amount of fossil fuel cars in the industry, as compared to the DfT's stats.

And the reason for that should be obvious from the two tables that will follow later this weekend - the TP figures grossly overstate numbers of hybrids and EVs, at least compared to the DfT's stats. But, of course, even ignoring TP's claim that one third of the cars in the NI combined trades are EVs, or that one quarter of the cars in the combined Scottish trades are EVs, it should have been obvious from the earlier tables that TP overstated the amount of EVs everywhere

But the corollary of that is evident from the table above - TP understates fossil fuel vehicles.

But note one oddball above. And indeed that's the one that initially drew my attention to the the fossil fuels category, and why perhaps it was worth digging into a bit more. Thus according to TP, Yorkshire & Humber has the highest proportion of fossil fueled cars in the whole of England at 68.3%. And by a significant margin - the next highest is the South East at 64.2%, and all the rest are in the 50s% or lower.

Yet according to the DfT stats, the result is totally the opposite - Yorkshire and Humber has the lowest proportion of fossil fueled cars in the whole of England and Wales.

And the reason for the DfT's stat should be obvious - the proportion of fossil fueled HCs is unremarkable, and it's the very low proportion of fossil fueled PHVs that's pulling the combined Yorkshire & Humber figure down - and because it's a weighted average, and only 14% of cars in Yorkshire & Humber are HCs, then the unremarkable figure of 80.2% for fossil fueled HCs doesn't pull the weighted average down much, so that figure equates to 42.2%.

And digging down a bit further, the reason for Yorkshire & Humber's exceptionally low figure for fossil fueled PHVs is obvious from the extract from the DfT's PHV table below:



So in Yorkshire & Humber the proportion of petrol PHVs is tiny at 2.7%, according to DfT, but it's really the 33.3% diesel PHVs that's pulling the fossil fuels figure down, because that's so much lower than all the other provincial regions. Only the West Midlands comes anywhere near at 44.1%, and the rest are generally up at around 60-70% or so.

And, of course, the other side of that coin is that the DfT's 'hybrid electric' figure for Yorkshire and Humber PHVs is easily the highest in provincial England at nearly 60% (in fact, even higher than in London, but of course the capital is exceptional in terms of plug-ins (both hybrids and EVs), as the table above illustrates.

But, to sum up, and in general terms, the first table shows that the TP numbers for the proportion of fossil fuels cars in the trade are significantly lower than the DfT's. Except for the region in which the DfT thinks fossil fuel cars are the lowest in provincial England - Yorkshire & Humberside. The same area that TP thinks has the highest proportion of fossil fuel cars

No, me neither