Location of Annie Millwood's attack

[Sam Flynn]

But still, if you're just trying to draw attention away, why then risk always going back to the same location if there are opportunities elsewhere?

Indeed, especially when the police presence in Whitechapel was massively boosted with reinforcements. Under such circumstances, why not kill in an adjacent district where police and civilian vigilance would have been significantly lower, and potential victims more readily caught off their guard?

[JeffHamm]

Thanks, Jeff. One problem with "commuters" in this context is that there were parts of London within a suspect's reach where potential victims (poor female vagrants, whether prostitutes or not) could easily be found, without his having to commute to Whitechapel. This was as true for Lechmere as it was for other putative commuters like Oswald Puckridge and William Bury. Quite how we'd incorporate potential killing zones outside Whitechapel into the model I don't know, but it's arguably academic in any case, as no Ripper murders were committed outside the small area already covered by the maps above.

Hi Sam,

I suppose one could argue that commuting is to draw attention away from one's home location. For example, David Berkowitz (sp: Son of Sam) commuted to the area where his crimes were, but surely there were couples closer to home? But still, if you're just trying to draw attention away, why then risk always going back to the same location if there are opportunities elsewhere? Whitechapel and surrounds was the poorest of the poor, and I suspect JtR was local lived in the area, and while he might have appeared to have the means for a knee trembler in Whitechapel, but he probably wasn't going to be looking too upmarket in other areas.

- Jeff

[Sam Flynn]

Ok, thanks Sam. Suspects well off the map, particularly east and/or west, would be "commuters", and geoprofiling starts by assuming the offender isn't a commuter (they are the minority). That's actually one of the questions I'm hoping to find an answer for, can we detect commuters? If so, can we build a separate algorithm that helps profile them? Even if we can't detect them, is there an algorithm for their spatial patterns (then one just produces both "maps", and says "this one is 80% likely the one you want, but around 20% of the time it's more like this - and if nothing is being found in the marauder zones at some point one might want to start looking in the "other zone 1, and so on".

Thanks, Jeff. One problem with "commuters" in this context is that there were parts of London within a suspect's reach where potential victims (poor female vagrants, whether prostitutes or not) could easily be found, without his having to commute to Whitechapel. This was as true for Lechmere as it was for other putative commuters like Oswald Puckridge and William Bury. Quite how we'd incorporate potential killing zones outside Whitechapel into the model I don't know, but it's arguably academic in any case, as no Ripper murders were committed outside the small area already covered by the maps above.

[JeffHamm]

Lechmere's home can't be placed on the same map; he lived further east.

Ok, thanks Sam. Suspects well off the map, particularly east and/or west, would be "commuters", and geoprofiling starts by assuming the offender isn't a commuter (they are the minority). That's actually one of the questions I'm hoping to find an answer for, can we detect commuters? If so, can we build a separate algorithm that helps profile them? Even if we can't detect them, is there an algorithm for their spatial patterns (then one just produces both "maps", and says "this one is 80% likely the one you want, but around 20% of the time it's more like this - and if nothing is being found in the marauder zones at some point one might want to start looking in the "other zone 1, and so on".

[Sam Flynn]

If you know where Grainger, Puckridge, and/or Lechmere were located, it would be great if you could stick some markers on the map.

Lechmere's home can't be placed on the same map; he lived further east.

[JeffHamm]

Puckridge was out of the asylum and living on Rupert Street, Soho in the West End. He was cleared by the City of London Police, who determined he had an alibi "beyond any doubt."

Thanks RJPalmer! Even if he hadn't been cleared, he would be a commuter, and as such, would mean one of the 20% or so of cases where one of the starting premises is wrong: which is that the offender is one of the 80%, a marauder. Also, his location isn't on the map, so I couldn't place him.

- Jeff

[rjpalmer]

Puckridge was out of the asylum and living on Rupert Street, Soho in the West End. He was cleared by the City of London Police, who determined he had an alibi "beyond any doubt."

PS. Personally, I'd consider Bachert a more likely "person of interest" than some of those mentioned. But carry on.

[JeffHamm]

hi Jeff
I don't think you need to toss out chapman (I use this name as its most common way hes referred as and avoids confusion with Koz). I think it would be safe to go with cable street which is close. the ones you need to toss out is bachert and pizer and ludwig. for future interest I would add valid suspects William grant grainger, Oswald puckridge, and Lechmere. I guess bury would count as a commuter killer, due to how far away he lived, so not included?

also, for simplicity just stick with the C5. smith should not be included at all. and where as I believe millwood, tabram and McKenzie were ripper victims, its too controversial. the biggest consensus is at least the C5.

Hi Abby Normal,

Just to clarify, while Smith is indicated on the map, that location is not included in any of the analyses (I can't see her as part of the series). As for the suspects, while I know some are cleared (Pizer had an alibi and was cleared at the inquest, Ludwig I was in custody during the double event, Barnett was interviewed for some time and apparently had an alibi that satisfied the police) and so forth; I recognize not all agree that they were cleared, however. I don't know much about Levy, just came across him as a suspect here and stuck him on the map. In part, and probably more useful than the analysis, is to create a single map with as many of the murder locations (C5 and others) and suspects (cleared or not), as a reference for everyone. I guess, iwth that in mind, you're right and Klowoski/Chapman's Cable Street location should be marked, and people can then discuss whether or not he was there at the appropriate time, etc.

If you know where Grainger, Puckridge, and/or Lechmere were located, it would be great if you could stick some markers on the map and post it then I'll update as well and include the zones for them. That would be much appreciated if it's not too much trouble (even more if it is, of course).

- Jeff

Edit: The short blub on Puckridge here on Casebook (https://www.casebook.org/ripper_medi...orley/152.html) seems to imply he might have been in an asylum from early 1888 until found later in the 1890s, but the information is very incomplete and he could have been released later in 1888. I'll have a more complete look when I get the time.

[JeffHamm]

Good work, Jeff! If I have time I will read this thread carefully. His route on the night of the Double Event is particularly interesting. Woolf Abrahams (brother of Aaron Kozminski) lived in Providence Street at the time of the murders, the Stride crime scene was around the corner (and 1882 he lived next to the Dutfields Yard). The Ripperīs route led from Dutfields Yard to Mitre Square at that night, quite possible he walked along Aldgate High Street/ Butchers Row where CID officer Sagar (not PC) watched a certain suspect afterwards. One might think that Sagar had watched "Kosminski" in Butchers Row, where this suspect had worked as a butcher. In this case and in this night it would have been possible that "Kosminski" walked from Butchers Row to Berner Street and back again, the places where he lived and/or worked. Therefore: You could find "Kosminski" and his victims near his home ("Berner Street/ Stride") and his workplace ("Mitre Square/ Eddowes") before 2 a.m. And after 2 a.m. till the small hours you could find him and his victims (Nichols, Chapman) in a "greater distance" from his home and/or workplace. Maybe it took some time in each night before the Ripper was ready to enter the northern area (above the High Street).

Karsten.

Thanks. Sure, that could work. In the end, though, we can always find some creative way to get a suspect where we need them to be (Sickert, Maybrick, Druitt, etc are prime examples). And, given the relatively small area, all walking distance, if JtR is just prowling around looking for opportunity, then anywhere in the general vicinity is really close enough to get to any of the offense locations (all the listed POIs above are close enough). My thoughts on being near Chapman (and perhaps Kelly too based on some accounts suggesting the possibility of the killer leaving relatively close to dawn) due to the time of day when JtR might have been on the move, does suggest to me a closer proximity to those locations than some of the others. But that's just me, and I could very well be wrong on that.

- Jeff

[Abby Normal]

Thanks. Ok, so I've updated the map to remove Klosowsky (which I somehow mis-spelled as Kowolsky above; sigh, it was late here) as Wolf V. points out he probably wasn't at that location in 1888, and have added PC Sagar's suspect in Butcher's Row, which appears to start at Middlesex Street and run westerly a short distance. So here's the updated list of zones (from highest interest to lowest based on C5 analysis):

Suspect Zone: ....... (C5/C5+Millwood and Tabram)
PC Sagar's Sus ...... ( 1 / 4)
Donston/Hospital ... ( 2 / 17)
Levy .................... ( 4 / 2)

Druitt* / Ludwig .... ( 7 / 14) Druitt is suggested to have had access to Dr. Thyne's surgery; Ludwig was ruled out as he was in custody on the double event
Barnet ................. ( 9 / 4)
Bachert ............... ( 9 / 13)

David Cohen ........ (16 / 17)
Tumblety ............. (16 / 33) There is some question as to whether he was here in 1888 or not
Hutchinson .......... (19 / 8)


Pizer ................... (32 / 40)
Kosminski ............ (36 / 47)
Kaminsky ............ (41 / 53)


There you have it, for what it's worth. With just the C5, PC Sagar's suspect in Butcher's Row appears to be in a location of the highest interest, where the search would start. The total area of each zone is 5% of the "better than chance" area (which I described earlier in the thread), and PC Sagar's suspect falls at the just under 2% mark (so well inside zone 1)

*Druitt was dead following Mary Kelly's murder, so if any of the later murders (McKenzie, Cole, etc) are part of the series, Druitt is out.

So, I'll include both the blank map, without Klowosky and with PC Sagar's suspect and the C5 output with the latter shown since this is the only suspect we've had so far to fall in zone 1. Again, this doesn't mean he was JtR - the purpose of geographical profiling is not to "solve" cases, that's for investigators to do based upon actual evidence (this isn't evidence!). Geographical profiling is just a tool that can be used to prioritize search strategies (if they are, in general better than chance) and may be useful in the prioritizing "persons of interest". If real evidence leads elsewhere, then the geoprofile's suggestion should be ignored in preference for actual leads and actual evidence. But, with limited resources of time and money, the idea of these sorts of things is to suggest where those resources may have a better than chance possibility of finding real leads.


and now the C5 based Geoprofile:



- Jeff

hi Jeff
I don't think you need to toss out chapman (I use this name as its most common way hes referred as and avoids confusion with Koz). I think it would be safe to go with cable street which is close. the ones you need to toss out is bachert and pizer and ludwig. for future interest I would add valid suspects William grant grainger, Oswald puckridge, and Lechmere. I guess bury would count as a commuter killer, due to how far away he lived, so not included?

also, for simplicity just stick with the C5. smith should not be included at all. and where as I believe millwood, tabram and McKenzie were ripper victims, its too controversial. the biggest consensus is at least the C5.

[S.Brett]

Good work, Jeff! If I have time I will read this thread carefully. His route on the night of the Double Event is particularly interesting. Woolf Abrahams (brother of Aaron Kozminski) lived in Providence Street at the time of the murders, the Stride crime scene was around the corner (and 1882 he lived next to the Dutfields Yard). The Ripperīs route led from Dutfields Yard to Mitre Square at that night, quite possible he walked along Aldgate High Street/ Butchers Row where CID officer Sagar (not PC) watched a certain suspect afterwards. One might think that Sagar had watched "Kosminski" in Butchers Row, where this suspect had worked as a butcher. In this case and in this night it would have been possible that "Kosminski" walked from Butchers Row to Berner Street and back again, the places where he lived and/or worked. Therefore: You could find "Kosminski" and his victims near his home ("Berner Street/ Stride") and his workplace ("Mitre Square/ Eddowes") before 2 a.m. And after 2 a.m. till the small hours you could find him and his victims (Nichols, Chapman) in a "greater distance" from his home and/or workplace. Maybe it took some time in each night before the Ripper was ready to enter the northern area (above the High Street).

Karsten.

[JeffHamm]

Now, because we have a list of suspects as well as the spatial analysis, the first 6 listed suspects would be recommended for closer inspection. In addition, despite Hutchinson's low zone (19), given his proximity to Chapman's murder, and non-spatial details we know of that crime (I'm thinking time of day here, as well as his known association with Kelly, and his insertion into the investigation), I would suggest over-riding the output of the spatial analysis and include him as a top-priority person of interest as well. Again, emphasizing that real information should over-ride a blind acceptance of a probabilistic based suggestion, which is what geoprofiles are.

- Jeff

[JeffHamm]

Take a look at here... located between Houndsditch and Middlesex Street or rather Minories and Somerset/ Mansell Street... not far from Mitre Square... https://www.casebook.org/dissertatio...tchersrow.html

Thanks. Ok, so I've updated the map to remove Klosowsky (which I somehow mis-spelled as Kowolsky above; sigh, it was late here) as Wolf V. points out he probably wasn't at that location in 1888, and have added PC Sagar's suspect in Butcher's Row, which appears to start at Middlesex Street and run westerly a short distance. So here's the updated list of zones (from highest interest to lowest based on C5 analysis):

Suspect Zone: ....... (C5/C5+Millwood and Tabram)
PC Sagar's Sus ...... ( 1 / 4)
Donston/Hospital ... ( 2 / 17)
Levy .................... ( 4 / 2)

Druitt* / Ludwig .... ( 7 / 14) Druitt is suggested to have had access to Dr. Thyne's surgery; Ludwig was ruled out as he was in custody on the double event
Barnet ................. ( 9 / 4)
Bachert ............... ( 9 / 13)

David Cohen ........ (16 / 17)
Tumblety ............. (16 / 33) There is some question as to whether he was here in 1888 or not
Hutchinson .......... (19 / 8)


Pizer ................... (32 / 40)
Kosminski ............ (36 / 47)
Kaminsky ............ (41 / 53)


There you have it, for what it's worth. With just the C5, PC Sagar's suspect in Butcher's Row appears to be in a location of the highest interest, where the search would start. The total area of each zone is 5% of the "better than chance" area (which I described earlier in the thread), and PC Sagar's suspect falls at the just under 2% mark (so well inside zone 1)

*Druitt was dead following Mary Kelly's murder, so if any of the later murders (McKenzie, Cole, etc) are part of the series, Druitt is out.

So, I'll include both the blank map, without Klowosky and with PC Sagar's suspect and the C5 output with the latter shown since this is the only suspect we've had so far to fall in zone 1. Again, this doesn't mean he was JtR - the purpose of geographical profiling is not to "solve" cases, that's for investigators to do based upon actual evidence (this isn't evidence!). Geographical profiling is just a tool that can be used to prioritize search strategies (if they are, in general better than chance) and may be useful in the prioritizing "persons of interest". If real evidence leads elsewhere, then the geoprofile's suggestion should be ignored in preference for actual leads and actual evidence. But, with limited resources of time and money, the idea of these sorts of things is to suggest where those resources may have a better than chance possibility of finding real leads.


and now the C5 based Geoprofile:



- Jeff

[S.Brett]

Interesting, do you know exactly where on the map they would be? Aldgate High Street is what Whitechapel Road turns into somewhere after Commercial Street I think, and it tends to all be low zone areas, but I would like to include them on the map above if you know where they were specifically.

Take a look at here... located between Houndsditch and Middlesex Street or rather Minories and Somerset/ Mansell Street... not far from Mitre Square... https://www.casebook.org/dissertatio...tchersrow.html

[JeffHamm]

You are both right. His "earliest known address" was 70 West India Dock Road in Poplar. He then moved to 126 Cable Street (where he was likely living at the time of the murders).

Wolf.

Ah, ok. Let's drop him.