An Expedia Car Rental Gone Wrong

I’ve written many times about issues with car rentals. Here’s another incident. Through Expedia, I booked a car rental from Sixt. My original reservation was for pickup on a Sunday at 5 PM with return on Friday at 2 PM, a 5-day rental. Due to a change in travel plans, I wanted to move the pickup to Monday at 9 PM while keeping the same Friday return, making it a 4-day rental.

I first tried calling Sixt and was able to speak to a live representative. I was told that because my booking was made through Expedia, they were unable to alter my reservation, although the representative left a note on my reservation that I had requested a pickup date on Monday instead of Sunday. I was instructed to call Expedia and ask to speak to their “Sixt internal account manager” who could make the change and also tell me if there was any price difference.

I then attempted to contact Expedia, but nowhere could I find a phone number to call. The only option I found was to invoke their Virtual Agent via chat. We sent messages back and forth about my request to change my pickup date from Sunday to Monday. It felt to me that the agent was assigned to chats with multiple customers simultaneously based on the time delays in the agent’s responses. In any case, I was eventually told that my only options were to keep the Sunday pickup or to cancel my non-refundable reservation which I had already paid in full. I already knew that my reservation was non-refundable, but I wasn't trying to cancel it. Furthermore, I did not see any language during booking or on my booking confirmation that no changes could be made. I was willing to pay the full amount I had already paid for five days even though I would only be renting the car for four, and I wasn’t asking Expedia to refund the unused day.

Sensing an unwillingness to help me, I asked the Expedia virtual agent to connect me with their “Sixt internal account manager” by typing into the chat the exact language used by the Sixt representative. The virtual agent eventually agreed to do so. After a long wait, I was told by the Expedia virtual agent that he was unable to reach the Sixt internal account manager or a Sixt representative and that he would have to send an email to Sixt. He told me to expect an email reply from Expedia in 48-72 hours. Right on cue about 2-3 days later, I received the following email from Expedia:

From: travel_expedia@expedia.com (travel_expedia@expedia.com)
To: (me)
Date: Sunday, September 13, 2026 at 09:45 PM PDT
Subject: Update on Your Refund Request with Sixt – Itinerary #xxxxxxxxxx

Hello Victor,

We’re following up regarding your request to cancel your booking and request a refund from Sixt.

Request type: Refund
Itinerary number: xxxxxxxxxx
Property name: Sixt

We spoke with Sixt to advocate for your refund request. Unfortunately, they have advised that the reservation was booked under a non-refundable rate, and they are unable to make an exception to their cancellation policy. As a result, no changes can be done for this booking.

Additionally, Sixt has informed us that the main driver named on the reservation must contact Sixt directly regarding any requested modifications to reservation xxxxxxxxxx.

We understand this is disappointing. We always strive to advocate for our travelers; however, Sixt has elected to uphold the terms and conditions associated with the reservation, and we are unable to override their decision.

If you have any questions or concerns, please feel free to reply to this email. We're here to help.

Regards,
Ipsita
Expedia US Travel Team

I can’t say for sure, but the response from Expedia felt inaccurate and potentially dishonest to me. The inaccuracy is that I never asked for a cancellation or refund. I had only asked to pick up my car a day later than originally scheduled. The potential dishonesty is that when I spoke to the Sixt representative, they seemed willing to allow the change and even left a note on my reservation stating that I had desired to pick up the car a day later. So it felt to me, based on my interactions with Sixt and Expedia, that Expedia was the one refusing to make the change, and they blamed Sixt. Lastly, Expedia wrote, “Sixt has informed us that the main driver named on the reservation must contact Sixt directly…” when in fact I had already contacted Sixt directly prior to engaging with the Expedia virtual agent. It seemed to me that Expedia may not have even contacted Sixt and simply gave me a boilerplate denial. Besides, how would the virtual agent know that emailing Sixt, receiving a reply, and responding to me would take 48-72 hours? I do not have definitive evidence for this, but my observations suggest to me that Expedia refused to accommodate a change and blamed their business partner for their decision.

I kept my existing reservation for a Sunday 5 PM pickup. I showed up to Sixt on Monday at 9 PM, and I asked if they would accommodate my rental. They did in fact give me the car, with a caveat. They explained that if a customer does not pick up their vehicle within 8 hours of their scheduled pickup time, they release the reservation. However, without much hassle, they did note that I had already paid in full, and the agent said he did his best to give me a new 4-day rental and credit me for the amount that I already paid. The best deal he could find for a similar vehicle type was about $73 more than I had already paid, and I thought that was a fair price given the situation.

In summary, Sixt seemed willing to accommodate my change when I called them on the phone, and they accommodated my change when I showed up in person. On the other hand, Expedia was uncooperative at best, contrary to their claim that they “always strive to advocate for our travelers”. It also felt to me that Expedia couldn’t get its facts straight and was possibly dishonest to me, as well as bordering on slanderous to Sixt. This experience has eroded my trust in Expedia.

Downloadable National Park Maps

If you’ve planned visits to US national parks, you probably know that each National Park Service (NPS) site has its own dedicated page or sub-site. Each page typically has a “Plan Your Visit” menu item where you can select from choice such as Basic Information, Things To Do, and Directions & Transportation. As I plan my visit, one of the first things I do is to view maps of the national park to understand the location of the visitor center(s), trails, and other features. In the past decade or so, various NPS sub-sites began transitioning their online maps from static downloadable PDF or JPG maps to interactive web maps that allow users to pan, zoom, and click on map features within their browsers.

If you’re old school like me, you might prefer downloading static maps, printing them out, and keeping a hardcopy in your backpack. Interactive maps are convenient while at home, but downloadable maps have a major advantage: they work when you don't have connectivity which is often the case in our national parks. If the NPS site does not provide downloadable maps, check out these alternatives.

National Park Maps is a passion project by a guy named Matt who has meticulously organized his website so users can look up maps alphabetically or by state. It’s an unofficial collection of free maps. I’ve often found maps on this site that are not even featured on corresponding NPS sites. Kudos to Matt!

NPS has its own official GIS, Cartography & Mapping site where you can download maps in PDF, JPG, or Adobe print production file formats. They can be searched by keyword, state, or park. Rather than providing the static maps in a separate website, it would be nice if each NPS sub-site offered options to view both interactive web maps and to view and download static PDF/JPG maps.

If you’re interested in learning more about the technology behind the interactive web maps, there is a National Park Service Web Maps site for that too. They discuss their open source NPMapJS code library.

While NPS has embraced interactive web maps, you don’t have to give up the good old fashioned downloadable map. Whether you prefer exploring a map on your screen, downloading a PDF for offline use, or stuffing a printed map into your backpack, there are still plenty of options. Happy mapping and happy exploring!

DROP Service Interim Update

Earlier this year, I wrote about how your can protect your privacy using California’s Delete Request and Opt-out Platform (DROP) service. DROP launched on January 1, 2026, and data brokers began processing requests on August 1, 2026. Given that data brokers are required to delete data within 45 days of receiving a request, I wanted to see how things were coming along at day 29. Here are my current stats, as reported on my DROP account:

And here is the DROP website’s glossary of deletion statuses:

When browsing the full list of data brokers, I see this:

Unfortunately there are 66 pages to scroll through on my account, and there does not appear to be a way to export the full list of data brokers along with my request statuses. It would be nice for them to add that feature later.

The DROP website provides the following information: “Data brokers will begin accessing DROP soon. Processing becomes mandatory beginning August 1, 2026, but status updates can take up to 90 days to appear in DROP depending on data broker workflows. Your status for each data broker will update once your request is processed.” My interpretation is that either (1) a majority of data brokers (490 out of 656) have not processed my deletion request yet and/or (2) some of those 490 data brokers have actually processed my deletion request, but the statuses have not yet been updated in DROP due to data broker workflows.

While the DROP website says, “Under law, data brokers have 45 days to delete your data after they receive your request,” it seems hard to determine if the law has been broken if the statuses can take 90 days to be updated. In the worst case scenario, a data broker could delete my data on day 45, and then it could take up to 90 days to update the DROP website, so that would be 135 days after August 1 which would be December 14, 2026. Maybe I’ll check back again toward the end of the year to see how well data brokers are complying with DROP requests.

University of California Acceptance Rate Data

If you have a high school senior who is getting ready to apply to University of California schools, you may have seen acceptance rate statistics such as the ones from Money Magazine a few months ago which I wrote about in my “LLM Analysis of College Ratings” blog. For convenience, I am re-posting the graphic here.

The acceptance rate statistics are from a nationwide sample, but what if I told you that you could look up University of California acceptance rate statistics for students at a specific high school? The University of California Information Center publishes data on Admissions by Source School and segments data into several subcategories as shown here:

The data tables show applicants, admissions, and enrollments by California high schools and community colleges for freshmen and transfer entrants by source school, mean GPA, ethnicity and school. Of particular interest might be “Single high school lookup” under the “Freshman admission by source high school” section. Data tables are provided for:

• Counts of fall freshmen by race/ethnicity
• Counts of fall freshmen by gender
• Average freshman GPA

After clicking on any of these 3 options, you will need to enter the high school name in the “Search for school name” field. Depending on the name of your school, the exact high school name could be tricky to find. If you come up blank, then go back to the home page and click on any of the 3 options under Freshman admission by source high school > All source high schools. Then filter by city and/or county and browse for the name of the high school of interest.

Back to the “Single high school lookup” section… After selecting a data table for counts by race/ethnicity, gender, or GPA, you will have the option to filter by year. You’ll also see breakdowns by campus, fall term, and applications/admissions/enrollment.

If you prefer to sort and filter data in Excel like I do, you’re in luck because there is a download button that gives you the option to download as an image, data, cross tab, PDF, or PowerPoint. Selecting “data” will result in the download of a .csv file which you can then play with in your favorite spreadsheet or database application.

I found this data to be useful because although the national acceptance rate statistics are good to know, being able to see acceptance rate statistics from my child’s high school provided a slightly different perspective. I hope you’ll find it as useful as I did. And if you found this data interesting, there are plenty of other data tables to explore at the University of California Information Center.

Citi Costco Visa Identity Verification

Last year our family visited several Asian countries, and we managed to get by with eSIM data-only plans. The only setbacks occurred when our Citi Costco Visa card flagged some transactions as suspicious (which I wrote about here and here), and because Citi attempted to notify me via voice and SMS text—which we did not receive—it prevented our purchases from being made. This year, I made sure to add my Google Voice number to my Citi Costco Visa card in hopes that I would be given an option to verify transactions using a data-only eSIM plan. It worked. When making a transaction, I was presented with the following options for identity verification:

You may have noticed that there were options to call or text me at the …2505 number, the the only option for the …2516 (Google Voice) number was to call me. I am not certain, but my phone contact information on the Citi website are:

• Home Number: …2505  → I think Citi uses this for calls
• Primary Mobile: …2505  → I think Citi uses this for text messages
• Alerts Mobile: …2516  → I think Citi uses this for calls

If my assumptions above are correct, then I think Citi should also present an option to send text messages to the alerts mobile number. While calling me at my Google Voice number worked for identity verification, I would have preferred getting a text message at my Google Voice number so I could copy and paste the code.

In any case, if you have a Google Voice or another VoIP (Voice over Internet Protocol) service, consider adding that contact number to your Citi card so that you can easily verify your identity using data-only eSIM plans when traveling abroad. I assume that this approach may work for other credit cards, but you’ll need to review your options and/or ask customer support.

EV Recommendations to Car Rental Companies

I have recently experienced several suboptimal car rentals which I’ve written about in this series. While traveling this week, I booked a rental car with Budget. The best deals were with electric vehicles (EVs), and given that I am comfortable with driving EVs and charging EVs at the company apartment, I chose an EV.

Upon vehicle pickup, I learned that Budget no longer provides charging cables. However, I discovered this in a circuitous manner. I recalled a prior experience in which I was given an EV with a low battery and without a charging cable. Therefore, I made sure to search my rental car for a charging cable—it didn’t have one. So I went back to the rental counter, and they assigned a different EV to me—that one didn’t have a charging cable either. I was then told by the manager that Budget no longer provides charging cables “because people steal them”. For context, not all car rental companies have this policy. Just last week, a co-worker rented an EV from another company, and it provided a charging cable.

When I rent EVs for my business trips to company headquarters, I always charge them at the company apartment which has electrical outlets in the 2-car garage, and for that I need a charging cable. The great benefit is that no time is lost by simply plugging in the car at the apartment. The alternative is to charge the vehicle at a public charging station which is undesirable for me because a typical level 2 public charger can take several hours to charge an EV from 50% to 80% capacity. So if a car rental company does not provide a charging cable along with an EV rental, that significantly worsens the customer experience.

I ended up switching from an EV to a hybrid vehicle, and I had a flawless experience with it. That being said, I have a few recommendations to all car rental companies who rent EVs to their customers.

First, car rental companies have a responsibility provide charging cables with EV rentals. Not every customer is willing or able to spend hours of precious time at a charging station. Business travelers are busy. Leisure travelers want to maximize their time on vacation. Nobody wants to use a public charging station unless it is their last option. Plugging an EV into a standard outlet is much slower than a level 2 charging station, but if it can be left unattended (e.g., while sleeping), the convenience far outweighs the slower charging speed. A charging cable should be considered a standard item included in an EV rental.

Second, charge the customer if they do not return the charging cable. If a customer borrows a GPS or a child car seat and fails to return it, then I assume the customer would be charged for those items. A car rental company should do the same for charging cables, and I would completely understand if the replacement fee for a charging cable exceeded the cost of a brand new charging cable because of the effort that must be taken to manage an inventory of charging cables. Furthermore, car rental companies already place a temporary charge on the customer’s credit card (Budget charged me $250 for this week’s rental), so they are already pre-authorized to charge fees for lost or stolen charging cables.

Third, charge the EV to 80% before letting a customer borrow it. This is a common customer expectation. It is also a common expectation that vehicles are adequately prepared for rental. The tire pressure should be adequate, the headlights should be working, none of the warning lights should be flashing, and the vehicle should be free of trash from the prior rental. Most car rental companies that I’ve used also require that the car be returned with at least 70% (or similar) charge, or a fee will be incurred (similar to a refueling fee for gas vehicles). On the flip side, car rental companies have a duty to provide customers with EVs whose batteries are 70-80% charged.

If a rental company cannot support EV rentals with the same level of convenience and preparedness that customers expect from gasoline vehicles, it shouldn’t offer them.

LLM Analysis of College Ratings

Money Magazine recently published research on the best colleges in America. Its methodology includes metrics that are weighted based on graduation rates, costs, financial aid, student debt, and alumni salaries. My daughter is looking at University of California schools (among other options), and I noticed that while UC Berkeley, UC Davis, UC Irvine, and UC San Diego achieved 5-star ratings, UCLA “only” received 4.5 stars.

Beyond what was stated in the methodology, I could not find detailed information about the raw scores that led to the star ratings, so I decided to consult various large language models (LLMs) to see if they could do some digging for me and propose plausible explanations for why UCLA wasn’t given a full 5 stars as I would have expected.

I asked the exact same question to Copilot, ChatGPT, and Claude: “Money Magazine released their 2026 ratings of the best colleges in the US (https://money.com/best-colleges/). Try your best to find out why UC Berkeley, UC Davis, UC Irvine, and UC San Diego achieved 5-star ratings while UCLA only received 4.5 stars. Make sure to examing their scoring methodology and determine what criteria may have caused UCLA to not achieve a full 5 stars.” Note that I copied and pasted the same typo (“examing” should have been “examine”).

Copilot didn’t spend much time researching this topic, and it started spitting out an answer 1-2 seconds after I pressed the “return” key. One of its main explanations for UCLA not achieving 5 stars was due to affordability which didn’t make sense based on the table above, so I followed up with a 2nd prompt: “UC Berkeley has a higher estimated full price and estimated priced with average aid than UCLA, yet UC Berkeley achieved 5 stars. Are you still confident in your assessment? If not sure, then state so.” Copilot then backtracked and revised its explanation, and its revision sounded more plausible to me. Here is Copilot’s full response.

A similar thing happened with ChatGPT which spent only a second or two longer to “think” than Copilot, and it responded almost immediately. It too explained that cost was a factor in it not achieving 5 stars, so I followed up with the exact same 2nd prompt: “UC Berkeley has a higher estimated full price and estimated priced with average aid than UCLA, yet UC Berkeley achieved 5 stars. Are you still confident in your assessment? If not sure, then state so.” Like Copilot, ChatGPT also backtracked and revised its explanation. Here is ChatGPT’s full response.

On the other hand, Claude thought long and hard about its response. It displayed multiple websites that it was consulting to formulate its response, including Money Magazine’s methodology page and the websites from each of the schools mentioned. I can no longer see the pages it was consulting, as they disappeared and were not displaying in the final response. Although I didn’t use a stopwatch, I’d estimate that Claude took 20-30 seconds to think before providing a response. I felt that Claude’s explanation was the most analytical and credible of the 3 LLMs, and unlike Copilot and ChatGPT, I didn’t feel that any of its responses were contradicted by the data. Here is Claude’s full response.

Of course, these are my personal opinions which are completely subjective. Also, my conclusion that Claude performed the best on this specific task are not necessarily generalizable to conversations about other subjects. The take home message is that if you often consult LLMs, I’d encourage you to always think critically about LLM replies and to explore multiple LLMs to compare and contrast their responses and determine which ones are best suited to your needs.