Forum Topic

My take is as follows.1. On safety we simply do not have enough data to make any reasonably statistically valid conclusion as to whether C9 is safer than prior to C9, although it does look as though the first iteration of C9 may have been less safe (again not enough data to be more definite than "may")2. The cycle km stat used by TfL looks to be statistical garbage.  The fact that they haven't explained how it is calculated is concerning until that data can be assessed for statistical soundness against what it is supposed to be assessing, it is probably best to ignore it completely.3. The number of cyclists using C9 has declined, but it is very unclear whether this is a temporary aberration or a longer term trend, probably need a couple more years of data to get any reasonably solidly based conclusion.  We know from other data that sales of bikes in UK has fallen dramatically (down 20%) but I suspect that is new bike sales and is failing to capture the rise of second hand bike sales (some of which are not stolen!) through gumtree, ebay etc and therefore may be exaggerating the decline in people buying bikes.  But even if the decline is 5% rather than 20% it would be consistent with data showing that the number of cyclists has declined.4. The pollution data needs careful analysis, there is a lot going on there and there are probably a multitude of different factors which need to be teased out to assess whether C9 has made any significant different either positively or negatively.5. I have issues with the data on bus times where TfL seem to be going out of their way to obscure the data trends.  In fairness some of that will not be their fault as the data has to be cleaned up to avoid the effect of the many days of roadworks which impacted on traffic speeds.  However, there are phrases in the report which are clearly designed to hide the fact that C9 has had a noticeable adverse effect on some bus journey times

Justin Stephenson ● 1096d

Justin,"Cyclist km" isn't statistical garbage - if you look at DfT road traffic statistics these are expressed in terms of vehicle km (or miles) https://roadtraffic.dft.gov.uk/summaryCalculating the risk of a collision uses vehicle km because if the number of collisions goes up, but the vehicle km goes up more, then the roads are actually getting safer as the collisions per vehicle km goes down.Having said that, I don't know why TfL didn't just keep it simple and just publish the counts at one of the survey locations and cyclist km can be used for other purposes where it makes sense.Regarding your claim "The number of cyclists using C9 has declined", there is no evidence of this.  The cyclist km increased, there has been no change to the length of Chiswick High Rd so the only way for the cyclist km to increase is the number of cyclists increasing.If you are willing to put your money where your mouth is, I'm happy to have a small wager that the number of cyclists will have increased between, say, June 22 to June 23 when either the TfL annual survey or camera survey publishes data.Air quality is impacted by a number of factors and I don't think anyone is claiming the improvements in air quality along CHR are solely because of C9. What the air quality data does show is that claims from opponents of C9 that air quality has got worse are nonsense.  Bike lanes meaning worse air quality has been a claim since day one and this has been comprehensively disproved but I expect like a zombie, it will continue to stumble on even though it is dead.

Michael Robinson ● 1096d

Justin, overall the collision injury risk has fallen taking into account the increase in cyclists but I agree that we don’t have enough data to claim that C9 is safer with statistical confidence: that will take several years of data. It is certainly true that incidents on the stretch from Chiswick Lane to Goldhawk Road increased with the first iteration but after the upgrade fell back. As Michael said driven or cycled km is a standard metric, but it it is difficult to compare to prior raw counts. The last data I’ve seen dates back to May 2022 before the gyratory cycle lane went in, so I’m not sure on what basis you can be confident that cycling has fallen. The hourly counts show that daytime cycling has increased very significantly, but commuting is still not back at prepandemic levels.  Certainly cycling London wide has fallen back from the peak of 2020, though by much less than the rest of the country.  Pollution is definitely down, as are vehicle counts. No surprise there. Bus times as a weighted average are up slightly, but peak time delays are lower, as shown in data from a previous FOI. I don’t think there’s been any attempt by TfL to hide these shifts; after all, they included the Chiswick Lane figures which showed the worst delays despite not being directly on the C9 route. That seems inconsistent with your claim that they “went out of their way to obscure the data trends”.The unattributed claim that there are less than 1000 cyclists a day on C9  stands out as the least credible comment made by anyone so far.

Tom Pike ● 1096d

"overall the collision injury risk has fallen taking into account the increase in cyclists"This seems to me a very dubious statement based on the available data. You can probably justify it if you accept the higher usage estimates based on 'cycle kms' and exclude the very high collision rates east of Goldhawk Road. Vehicle kms may be a standard metric but the way that TfL has calculated it for its report is quite innovative and it admits data has been excluded from the analysis that is 'challenging'. I still don't fully understand how the calculation is made but if it is showing an increase while at the same time more straightforward cycle counts are  showing a decline any growth numbers from this method should be pencilled in to an analysis not used to make bold claims on the matter of safety for cyclists.On bus delays, they clearly are a concern and you really shouldn't be trying to detach the longer journey times on Chiswick Lane from C9 because even TfL are making the association. It is interesting that the Monitoring Report chose not to use your claim that peak time delays are lower which would have significantly strengthen the argument it was trying to make.I'm uncomfortable with you trying to minimise safety issues and bus delays not because I am trying to make a case that C9 should be removed but because, the more you do this, the more it reduces pressure on TfL and Hounslow Council to act to mitigate these problems.On the Chiswick section of C9, the issue of safety remains very much a live one and TfL should not be seeking to minimise it. They have clearly been caught doing this by the call-in report which makes a convincing argument that more data is need before it can be said that the recent junction treatments have reduced risk to acceptable levels. That's why I think Hounslow should reconsider making the scheme permanent - enabling it under another experimental order would ensure the safety issue is re-examined after a reasonable period of operation.On buses, Hounslow Council, to its credit, does not seem to have accepted your argument that the impact on journey times is tolerable and is proposing quite radical changes to restore the quality and reliability of the service. This may prove to be expensive and the temptation to devote scarce resources elsewhere will increase if they have a steady supply of testimony from a respected professor that there isn't really a problem.

Jeremy Parkinson ● 1095d

Jeremy, the decision is on the Hounslow section and the monitoring report reflects that. The drop in injuries is apparent from the recorded bike counts, not the cycle-km numbers; as I think we agree, there has not been sufficient time to say that the rate has dropped by a statistically significant level. Where is the "bold claim" there? To quantify, it would take four or five more years at the current rate to be able to demonstrate that level of confidence, and very possibly rather longer. Are you suggesting that every relatively short length of cycle infrastructure should be regarded as temporary until that standard has been reached? In some cases that might take decades! Is that a "reasonable period of operation"?In fact we have evidence on the effect on average cycle injury rates by much more extensive studies based on hundreds of kms of cycle infrastructure in the Netherlands, Denmark and elsewhere as well as other studies that have looked at injury rates in London. Notably, there is good evidence in these studies of "safety in numbers", that as more people cycle in response to improved infrastructure, more general awareness of them by drivers means that the risk of injury collisions drops, which in turn encourages more people to cycle. That is part of the explanation as to why London has seen significantly higher growth in cycling rates than the rest of the country and has retained much more of the growth in cycling seen during the pandemic. It is impossible to discern these effects from just one mile of cycle lane. I am not trying to "detach" the Chiswick Lane bus running times, quite the opposite! And it is not a "claim" that westbound peak delays are lower on the High Road, that is simply what the previously released data showed;  it is not inconsistent with the weighted average showing a limited increase. Only TfL can calculate a weighed average as it requires data on passenger numbers which has not previously published. As for the "radical" changes proposed, they include a reversion to the retention of the bus lane closer to the original plans for C9, as is the redesign of the junction at Heathfield Terrace. I fully support such changes, but as you say, the issue will be the availability of funding.

Tom Pike ● 1095d