How far in advance to book?

[cpotisch]

Cpotisch is correct in what was said, above. However, I wouldn't let that keep me from using it. Although it can send spasmodic and nonsensical notifications, if it comes through with a correct notification of a fare drop once in a while it's worth establishing a Fare Watch. Who knows. . .

. . .that may be the day you would have forgotten to do a manual fare check. . .and maybe the fare would have gone back up some hours later . . . just before you finally remembered to do your daily manual fare check.

You can't always sometimes tell what you least expect to happen the most.

I just checked, and it looks like Paul finally shut down Fare Watch. He had said a few weeks ago that it was just too glitchy for it to be worth keeping out there, and now the page just says "under construction", so at least for the time being, Fare Watch is gone.

 

[cpotisch]

Of course this was only for 2 of the 32 possible LD trains with sleepers (both directions) so is but a small sample of the whole.

32 LD trains with sleepers? There are only 13.

1. Lake Shore Limited
2. Cardinal
3. Silver Star
4. Silver Meteor
5. Crescent
6. City of New Orleans
7. Capitol Limited
8. Texas Eagle
9. Sunset Limited
10. Empire Builder
11. California Zephyr
12. Southwest Chief
13. Coast Starlight

 

[niemi24s]

There seems to be a rule of thumb that the best time to book a sleeper is 2 to 5 months or maybe 3 to 5 months in advance of the trip in order to garner the lowest sleeper fares. Did a check of two LD trains for the total number of low bucket Roomettes available for each of the 11 months starting tomorrow and found (in chronological order): 9,38,32,16,18,24,28,30,34,39,42.

• As expected, the fewest low bucket Roomettes were available during the 30 days hence.

• The next fewest were available during the January & February periods

• The most low bucket Roomettes were available during the November, July and August periods

Of course this was only for 2 of the 32 possible LD trains with sleepers (both directions) so is but a small sample of the whole. But based on this small sample I'd venture a guess that the "best time" to look and book is 10 or 11 months out - or as soon as you know when you want to go. Methinks this much-touted rule of thumb has little merit. Others are cordially invited to use AmSnag to conduct their own checks and post their results - with data of course.

Which two trains?

Are you asking because you don't want to duplicate my efforts - or because you want to churn up reasons why my data is not representative of the whole (which I already alluded to)?

Whatever the reason, I'll not say simply because I fully intend to check for low bucket Roomettes out 11 months on all 32 numbered trains within the next week - some day when the weather is lousy. Had our first snow last night, so it won't be too long from now.

 

[Maglev]

I think niemi's 32 trains includes Auto Train, Palmetto and both sections of the Empire Builder.

I have been watching fares for a trip in January since I booked in May. I have been able to modify my fare on one leg to a lower bucket, and upgrade on another leg. Even though the whole trip is now at low bucket, I continue to watch the fares.

They make no sense, except that sometimes the count of rooms goes down and if it was one room available, the fare will go up. But when to find the best fare is anybody's guess.

 

[cpotisch]

There seems to be a rule of thumb that the best time to book a sleeper is 2 to 5 months or maybe 3 to 5 months in advance of the trip in order to garner the lowest sleeper fares. Did a check of two LD trains for the total number of low bucket Roomettes available for each of the 11 months starting tomorrow and found (in chronological order): 9,38,32,16,18,24,28,30,34,39,42.

• As expected, the fewest low bucket Roomettes were available during the 30 days hence.

• The next fewest were available during the January & February periods

• The most low bucket Roomettes were available during the November, July and August periods

Of course this was only for 2 of the 32 possible LD trains with sleepers (both directions) so is but a small sample of the whole. But based on this small sample I'd venture a guess that the "best time" to look and book is 10 or 11 months out - or as soon as you know when you want to go. Methinks this much-touted rule of thumb has little merit. Others are cordially invited to use AmSnag to conduct their own checks and post their results - with data of course.

Which two trains?

Are you asking because you don't want to duplicate my efforts - or because you want to churn up reasons why my data is not representative of the whole (which I already alluded to)?

Whatever the reason, I'll not say simply because I fully intend to check for low bucket Roomettes out 11 months on all 32 numbered trains within the next week - some day when the weather is lousy. Had our first snow last night, so it won't be too long from now.

Seriously? She literally just asked you which two trains you're talking about, and you immediately jumped to the conclusion that she likely wanted to spite you or prove you're wrong. Sometimes people just ask an honest question with no ulterior motive and which don't reserve some nonsense snarky remark about the question not needing to be answered.

 

[niemi24s]

[Redacted. Please see next post]

 

[niemi24s]

Finished gathering the data for all 32 numbered trains with sleepers. Thought the best way to present the data was by showing the average number of days Roomettes were available for all the trains - whether or not they had any available - for each of the 11 thirty day periods. This is, after all, a look at the merit of the rule of thumb about the best time to look for a sleeper so there was no need to show any data for the individual trains. Rules of thumb here are put forth as if they apply to any and all trains.



While the greatest average number of days Roomettes were available was during the 4th and 5th thirty day periods ("months") from the present it can be seen that there's no reason not to start looking for one out at the 11 month limit. But treat this info with caution so to not get misled. Here are some observations from looking at all 352 items of raw data:

• 12 of these 32 numbered trains had no Roomettes available during the last 4 to 6 months of this period. However,

• 3 had 22 or more Roomettes available during each of the last 6 months of this period and

• 9 had 22 or more Roomettes available during the very last (11th) month of this period.!

Conclusion: please don't be misled by those who say they only look for Roomettes within 5 months of travel. Start looking as soon as you want to - out to the 11 month limit. Even if it's at high bucket for a fixed travel date at least you can book it and be assured of having it. If the price drops, get the fare modified. If it never drops......

 

[cpotisch]

Conclusion: please don't be misled by those who say they only look for Roomettes within 5 months of travel. Start looking as soon as you want to - out to the 11 month limit. Even if it's at high bucket for your fixed travel date - at least you've booked it and are assured of having it - if the price drops, get your fare modified.

Nobody said that. We just said that in our experience, the low buckets tend to start showing up more at around the five month mark.

 

[niemi24s]

I wonder who could have possibly written this back in Post #2?

". . . you probably won't find any benefits of booking much before the five month mark."

 

[niemi24s]

Or the following from Post #4 in this thread: http://discuss.amtraktrains.com/index.php?/topic/73712-sleeper-prices-wow/

From what I've seen and heard, there's usually a price drop about five months in advance, so you'll often find the best prices around then. And when you think about it, when it's 11 months out, Amtrak is in no imminent danger of the room going unsold.

If you'd take the time to use AmSnag to examine sleeper price fluctuations from day to day, you'd see there are price drops and price increases all during the 11 month availability period and you wouldn't have to parrot that 5 month malarkey mentioned by others. But don't feel bad - I've heard the same malarkey from several other AU members over the years. Yours were just the most recent and, therefore, easiest to find.

For a majority of trains you can find the best (lowest) prices at some time during any of the 11 months of availability. F'rinstance, a low bucket Roomette on the SS is $187 and here's what AmSnag shows available as of about 10 minutes ago during the very, very last 30 days of the 11 month availability period:



Please don't ever tell me or anybody else there's little use in looking ahead 11 months for the best (lowest) fares. They're out there (depending on the train, of course). And there are (at least, were) 8 others with 22 to 30 low bucket Roomettes available during the last, 11th, month of availability. See for yourself. Get the facts. Please don't be an unwitting contributor to the malarkeyness of AU.

 

[cpotisch]

I wonder who could have possibly written this back in Post #2?

". . . you probably won't find any benefits of booking much before the five month mark."[/font

 

Yes, I know. I said that.

 

And it's the same thing that I said here:

 

Conclusion: please don't be misled by those who say they only look for Roomettes within 5 months of travel. Start looking as soon as you want to - out to the 11 month limit. Even if it's at high bucket for your fixed travel date - at least you've booked it and are assured of having it - if the price drops, get your fare modified.

Nobody said that. We just said that in our experience, the low buckets tend to start showing up more at around the five month mark. 

 

And the same thing that I said here:

It used to be that the cheapest fares were found the full 11 months in advance. Now from what I've seen and heard, the sweet spot is about five months in advance. It might vary between cases, but I would say all things being equal, you probably won't find any benefits of booking much before the five month mark.

You act like I've been contradicting myself, but I really haven't.

 

[pennyk]

Please note that personal attacks on other members violate AU's rules and guidelines. Posts containing such attacks may be removed.

 

[niemi24s]

Personal attack? YGTBSM! I was simply trying to explain to him with cold, hard facts how his advice was flawed.

OK, I concede there's really little use in looking out beyond 5 months.

Even though 47% (2004 of 4272) of the the days when low bucket Roomettes were available was during the next (last) 6 months of the 11 month period.

I give up.

P.S.: Second line is pure sarcasm - sorry.

 

[Ryan]

How did you determine how many roomettes are available on each train? Amsnag doesn’t show inventory levels.

 

[niemi24s]

I changed my mind about giving up. Our Rules and Guidelines states...

" Debate the value of ideas, not the value of the person expressing the ideas."

...so here are the facts as gleaned from many AmSnag searches conducted over a three day period, taking about six hours to gather and another couple of hours to carefully analyze. The idea being debated here is the one that asserts:

". . . you probably won't find any benefits of booking much before the five month mark."

My own assertion is that this idea has no value whatsoever when it is applied to all 32 numbered Amtrak LD trains. This assertion and its corollary (that there are indeed significant benefits in looking/booking beyond the five month mark) are supported by the following facts:

1. There are 32 individually numbered Amtrak trains with sleeping accommodations.

2. Fares are posted by Amtrak for the future 11 month period.

3. As AmSnag (thanks, Paul) can only gather and display fare data from Amtrak in at most 30 day increments, the complete search for Roomette upcharges required 11 X 32 = 352 individual AmSnag searches containing 352 X 30 = 10,560 lines of fare data to scrutinize for days with low bucket Roomettes.

4. Raw data (the number of days low bucket Roomettes were available) from each of these 352 searches showed that each 30 day period had a total number of days when low bucket Roomettes were available on all 32 trains of (in chronological order starting on either 5, 6 or 7 Oct 2018): 177, 434, 321, 646, 690, 290, 284, 313, 347, 347, 374 and 396.

5. The total number of days when low bucket Roomettes were available during this 11 month period was 4,272.

6. The total number of days when low bucket Roomettes were available during the first 5 months of this period was 2,268.

7. The total number of days when low bucket Roomettes were available during the last 6 months of this period was 2,004.

8. The percent of all days when low bucket Roomettes were available during the last 6 months was therefore (100 X 2,004) ÷ 4,272 = 46.9%

CONCLUSION:

The idea that...

". . . you probably won't find any benefits of booking much before the five month mark."

...has no merit because it ignores facts clearly showing that heeding it will lead to overlooking, ignoring or failing to consider about 47% of the days with the best (lowest) available Roomette fares - those during the last 6 months of this period.

 

[tricia]

To repeat Ryan's question: "How did you determine how many roomettes are available on each train? Amsnag doesn’t show inventory levels."

Or did you mean "number of days on which low-bucket roomettes are available" instead of, for example, "5. The total quantity of low bucket Roomettes available ...." ?

And I'm still curious about whether you noticed any different patterns of availability for different trains.

 

[niemi24s]

I didn't determine how many Roomettes were available on each train. But I can now see how my wording was simply awful and your "number of days on which low-bucket roomettes are available" is spot on. Thanks for that. Sometimes my tongue gets wrapped around my eye tooth and I can't see what I'm saying.

It'll take me a few hours to go back and edit the text and image.

Although the wording will change, the numbers will remain the same.

And yes, I did notice different patterns of availability for different trains:

• Some reached a maximum at the 4th/5th month and then fell to zero at month 6 or 7.

• Some reached a maximum at the 4th/5th month and then fell slightly and rose to a higher level by the 11th month

• A few increased gradually throughout the period and reached their maximas at the 11th month.

• Still others seemed to have no discernible pattern at all.

It's quite difficult for me to try to adequately describe in words the patterns seen in all 32 numbered trains. Two better ways to display it all would be by posting all the raw data in tabular form (all 352 numbers including their sums) or do the same by plotting them graphically. But let me tidy up my wording first and then see what I can do about the others.

 

[Ryan]

Thanks, that makes far more sense.

I'd love for you to share you data sets some day.

 

[niemi24s]

Thanks, that makes far more sense.

I'd love for you to share you data sets some day.

Yeah, I really got all wrapped around the axle with my wording of all this stuff. Thanks for questioning what I wrote and prompting the edits. Hope it all makes better sense now.

Gave the nonverbal presentation of this data a lot of thought and have initially concluded that any graphic portrayal would be a mess - what with a line for each of the 32 trains. And some routes have significantly different data between the travel directions and that would get masked if the data for each direction was combined.

So that leaves the tabular form of the raw data, and that'll take the rest of the evening for my able assistants (Sons) to make up a spread sheet with 32 rows and 11 columns. That's because at present it's all nothing more than a hand-scribed single sheet of lined notebook paper. Give me some time on this.

Thanks again!

Edit: Well, it didn't take as long as I thought it would. Here's all the tidied up raw data - have fun with it.

Edit #2: New chart in place of the previous one, below, with an added column on the right showing availability for each of the 32 trains. This allows better recognition of route direction differences - some of which seem quite large.

Edit #3: Replaced the hand-scribed chart with a spreadsheet/FastStone version. Data is unchanged.

 

[niemi24s]

Q: If you lived in CHI and wanted to take the SWC to LAX in a Roomette next August, would you start looking now or wait until March - 5 months before the trip?

<no need to reply>

 

[cpotisch]

Q: If you lived in CHI and wanted to take the SWC to LAX in a Roomette next August, would you start looking now or wait until March - 5 months before the trip?

<no need to reply>

I would take a look now, but if I don't see any low bucket fares this far out, I wouldn't freak out because they'll likely show up more as it got closer in.

 

[niemi24s]

Sometimes a table with 352 numbers can be a bit intimidating but a graph of it all may less so:



Maybe.

 

[Ronbo]

Sometimes a table with 352 numbers can be a bit intimidating but a graph of it all may less so:

[attachment=10636:LB Roomette 2D GraphA.jpg

Maybe.
[/QUOTE]From a relatively simple question from the original poster, I hope she got her answer, to this ️Amazingly ridiculous ️However, quite artistic, like looking at a Jackson Pollack painting

 

[Ryan]

Excellent graph.

Despite a handful of outliers, there's a pretty clear pattern of lots of low bucket availability in the 4-5 month window.

It would be interesting to repeat the analysis over time to determine if the abundance of low bucket rooms is because of the 4-5 month window, or because that's where Jan/Feb (a.k.a low travel season) falls. My money is on the latter.

 

[niemi24s]

From a relatively simple question from the original poster, I hope she got her answer, to this ️Amazingly ridiculous ️