How looking at reservations in isolation makes me overpay for travel

There is a mistake I consistently make in my travel hacking practice: I compare every travel purchase I make to its lowest-cost alternative in isolation. For example, take an airport like LAX with nearby properties in multiple hotel chains. Glancing at a site like Awardmapper, you can compare these locations:

  • Hilton Los Angeles Airport (32,000 — 40,000 points);
  • Hyatt Regency Los Angeles International Airport (12,000 points);
  • a couple of Wyndham properties (15,000 points);
  • Four Points by Sheraton Los Angeles International Airport (10,000 points);
  • and Holiday Inn Los Angeles - LAX Airport (30,000 points).

You can do a quick calculation based on your own travel hacking practice, but using my values I'd come up with a cost of $160, $120, $150, $200, and $210, respectively, as the imputed redemption value for each of those properties. That's not the out-of-pocket cost of acquiring the points, but rather the foregone rewards on a 3% cash back card in the case of Hilton, a 2% cash back card in the case of Wyndham and Sheraton, a 1 cent cash back redemption of Ultimate Rewards points in the case of Hyatt, or a 0.7 cent-per-point purchase of IHG Rewards points in the case of the Holiday Inn (although you can almost always do a bit better than that).

So far, so good. If the cost of a suitable hotel were less than $120, I'd book with cash, and if more, I'd book with points, in this case probably 12,000 Ultimate Rewards points transferred to World of Hyatt.

This makes me too hostile to rewards gimmicks

I've written plenty of times about how much I dislike the annual companion ticket that comes with my Delta Platinum Business American Express card. The problem with it is simple: you have to pay for the primary passenger's ticket (and the taxes and fees for the companion) with an American Express card, and that means paying cash. If you'd otherwise buy your ticket with US Bank Flexpoints, for instance, then a sub-50% discount (with only one ticket earning flight credit) just isn't that impressive compared to the roughly 58% discount I'd get on a pair of Flexpoint redemptions, both of which would be eligible for earning and upgrades. Adding in the $195 annual fee makes the companion ticket an even worse value.

Likewise I have no use for annual free night certificates at chains I wouldn't otherwise stay at, since the additional cost of staying more than one night would eat up any potential savings on the first night.

This is fine, as far as it goes, but is also the source of the mistake I'm talking about: I also have to pay with cash if I don't have enough points to redeem for my airline tickets, or hotel rooms, or car rentals, or whatever. Now, I can put the charge on my Arrival Plus card and get trip delay insurance, and maybe redeem $100 in rewards as a statement credit against the purchase sometime in the future, but that's no substitute for an up-front 58% discount.

Travel hacking ideally reduces your total out of pocket costs

That brings me to the point of this post, and what I'm sure is a better way to ultimately implement a travel hacking strategy than mine: getting the biggest discounts on the expenses that are hardest to hack, and accepting smaller discounts where necessary in order to reduce your overall travel costs.

Some stylized facts demonstrate this clearly:

  • Two $400 roundtrip tickets booked using a Delta Platinum companion certificate would cost $405.60 (depending on connections), plus a $195 annual fee, for a total of $600.60, a 25% discount.
  • The same two tickets would cost 53,333 Flexpoints, representing $26,667 in grocery store spend, costing roughly $337 in activation and liquidation fees, a 58% discount.

If your only travel expense during the year is buying two $400 Delta tickets, then you're obviously better off redeeming Flexpoints and realizing a 58% discount than using a companion ticket and saving just 25%. And indeed, this is my main argument against companion tickets that require you to pay cash.

The problem with this logic is that the 58% Flexperks travel discount can be applied to a whole range of travel expenses, now even including travel expenses made directly with the vendor under certain circumstances.

If you had unlimited access to manufactured spend, you would always have enough points to redeem and pay the lowest possible amount out of pocket. Whenever Hilton, or Hyatt, or IHG offered the largest discount, you'd book with those points, and book flights with miles, Flexpoints, Ultimate Rewards points, or whatever other currency offered the lowest out-of-pocket cost.

But if you have finite access to signup bonuses, manufactured spend, airline credit, etc., then misallocating those currencies can mean paying cash and realizing no discount at all on some expenses. To return to the example above, if in addition to your $800 in airfare you also need an $800 hotel room, then you'd be better off taking the 25% discount on your flights and applying the 58% discount to your hotel, instead of taking 58% off your flights and 0% off your hotel: $937 is less than $1,137.

Framed in this way the solution is obvious because both expenses are connected to the same trip. Over an entire year, that won't be the case: sometimes you'll fly American instead of Delta, you'll fly internationally instead of domestically, you'll have sub-$500 hotel stays that won't qualify for Real-Time Rewards redemptions, etc.

It's that inter-temporal coordination problem that causes my sloppy thinking and leads me to pay more than I would if I optimized my redemptions across longer planning periods.

Conclusion

I don't think there's any shame in being wrong, but you do tend to run into problems when you refuse to admit you're wrong, so I fully admit this is something I'm wrong about: I try to redeem miles and points as aggressively as possible, even when it leaves me holding the bag and paying cash for tickets or hotel rooms I could have gotten a better deal on if I sorted my redemptions by value ahead of time.

I'm so resistant to paying cash that I deliberately book my Delta companion ticket on the cheapest flights possible, even if that means redeeming far more Flexpoints or Ultimate Rewards points on more expensive flights in the future.

But you don't have to be as dumb as me! By planning ahead and creating even a crude demand schedule, you're much more likely to lower your out of pocket costs by using the right rewards tool for the right job.

Marriott Travel Packages are getting more expensive (also maybe more valuable)

I saw Spencer Howard post on Twitter a seemingly-official (or at least well-spoofed) document listing the new prices of Marriott Hotel + Air packages when the combined Marriott, Starwood, and Ritz-Carlton program goes into effect on August 1.

Let's talk about it.

Marriott Hotel + Air packages today

Today, Marriott Hotel + Air packages offer a discount of roughly 24 to 38% when redeeming Marriott Rewards points for 7 consecutive nights and a variable number of miles transferred to one of their partner airlines:

  • The smallest package offers 7 nights at a Category 1-5 property (worth up to 150,000 Marriott Rewards points since the 5th night is free in any case) and 50,000 miles, worth 120,000 Marriott Rewards points (since that amount can be converted into 40,000 Starpoints and transferred to many of the same airline partners, although the Starwood transfer ratio to United is much worse), but costs just 200,000 points, a 26% discount.
  • What seems to be the most popular package among travel hackers offers the same 150,000-point, 7-night stay plus 120,000 miles, worth 288,000 Marriott Rewards points using the same calculation above, but costs just 270,000 points, a 38% discount (giving rise to the odd situation discussed here).
  • Finally, the largest package offers a 7-night Tier 5 Ritz-Carlton stay, worth 420,000 Marriott Rewards points, plus the same 288,000-point mile transfer, but costs just 540,000 points, a 24% discount.

Again, these calculations are based on the current Hotel + Air award chart.

Marriott Hotel + Air packages after August 1, 2018

Now let's apply the same logic to the Hotel + Air chart Spencer posted for redemptions after August 1, 2018.

  • At the low end, you can redeem 255,000 points for a 7-night Category 4 stay plus 50,000 miles, which would otherwise cost 270,000 Marriott Rewards points: 150,000 points for the 7-night stay, and 120,000 points for the mileage transfer, a mere 5.6% discount.
  • The same stay certificate plus 100,000 miles costs 330,000 points, raising the discount to 15.4%.
  • At the high end, a 7-night Category 8 stay (starting in 2019) plus 50,000 miles costs 675,000 points and is worth 630,000 points, a 7% premium over making the two transactions separately!
  • Transferring 100,000 miles instead raises the value of the package to 750,000, which is, in fact, what the most expensive package costs.

This pattern repeats itself for the other packages as well: discounts are small or negative at the 50,000-point redemption level, and range from 5% to 15% at the 100,000-point level.

Marriott Hotel + Air Packages after January 1, 2019

While losing an opportunity to book hotel stays and buy airline miles with a 38% discount is unambiguously a devaluation, I think the explanation is not what will happen in August, but what will happen next January, when Off-Peak and Peak pricing goes into effect. These 7-night stays make no sense at almost category property during Off-Peak periods (an Off-Peak Category 6 redemption would cost just 360,000 points booked separately, but Marriott charges 415,000 points!).

But during peak periods, the discount can be noticeable even at Category 8 properties, where a 50,000-point package will cost 675,000 points but is worth 720,000 points (6.3% discount) and a 100,000-point package costs 750,000 points but is worth 840,000 points (10.7% discount). Note also that these Category 8 properties will include Starwood's current top-end properties in addition to Marriott's.

And if you book a Peak Category 4 hotel with a 100,000-point package, the discount rises to 21.4%, which is at least in the same ballpark as the existing packages.

However, these are still much lower discounts than those offered by the current Hotel + Air packages, so you should certainly book any packages you're interested in before the August 1 changes go into effect. There's been a lot of speculation about how unredeemed stay certificates will be treated after August 1, but given the discounts currently available I think it's largely irrelevant: getting back any points at all from unredeemed certificates will make them an even better deal, and updated points-based certificates would still be able to be used at the much larger joint Marriott-Starwood hotel footprint.

Not too many Ultimate Rewards points angles

Since I get any Marriott Rewards points I need through Ultimate Rewards transfers, I'd like there to be some way to take advantage of Hotel + Air packages that way, but the fundamental problem is that Ultimate Rewards points can already be transferred to programs in each airline alliance, Southwest, and Hyatt at a 1:1 ratio, while Marriott Rewards points can only be transferred at a 3:1.25 ratio.

That means while Hotel + Air packages are good redemptions of Marriott Rewards points (improving their value over individual stay redemptions and individual airline transfers), they're bad redemptions of Ultimate Rewards points.

The exception is if you already have a large Marriott Rewards balance you're considering transferring to Starwood Preferred Guest in order to make an airline transfer. In that case, you should consider instead transferring over Ultimate Rewards points in order to redeem a Hotel + Air Package. For example, 240,000 Marriott Rewards points are worth 80,000 Starpoints, which can be redeemed for 100,000 miles. But transferring 10,000 Ultimate Rewards points over to Marriott will allow you to receive the same number of miles (or more in the case of United MileagePlus), but also a 7-night Category 1-5 certificate.

No matter what Marriott decides to do with those certificates on August 1, it's virtually guaranteed to be worth more than 10,000 Ultimate Rewards points.

Transfer Starpoints to Amtrak Guest Rewards before August 1, 2018

For many years I was a booster of Chase Ultimate Rewards transfers to Amtrak Guest Rewards, due to their zone-based redemption system which made it possible to get 3 or more (possibly much more) cents per point when redeeming Amtrak Guest Rewards for transcontinental sleeper accommodations.

Unfortunately, in December, 2015, Ultimate Rewards points could no longer be transferred to Amtrak Guest Rewards, and in January, 2016, Amtrak moved from a zone-based to a fixed-value redemption scheme.

However, those fixed-value points are still quite valuable!

Refresher: the value of Amtrak Guest Rewards points

Amtrak Guest Rewards points are similar to Southwest Rapid Rewards points in that they have fixed values, but the value they're fixed at depends on the redemption in question. For example:

  • a coach seat on the Northeast Regional from Washington to Boston costs $79 or 3,830 points, for 2.06 cents per points;
  • a business class seat on the Acela Express between the same cities costs $138 or 7,176 points, for 1.92 cents per point;
  • a first class seat on the Acela Express costs $282 or 10,998 points, for 2.56 cents per point;
  • a coach seat between Chicago and Los Angeles on the Southwest Chief costs $142 or 6,107 points, for 2.33 cents per point;
  • a Superliner Roomette costs $794 or 27,393 points, for 2.9 cents per point;
  • a Family Bedroom costs $1,158 or 39,951 points, for 2.9 cents per point;
  • a Superliner Bedroom costs $1,606 or 55,407 points, for 2.9 cents per point.

Starpoints can be transferred to Amtrak Guest Rewards through July 31, 2018

Starwood has announced that transfers to Amtrak Guest Rewards will end with the introduction of the new Marriott Rewards program on August 1, 2018, although it's fair to speculate whether Marriott will arrive at their own accommodation with Amtrak after that date.

Amtrak redemptions are good and you should consider speculatively transferring points

Amtrak Guest Rewards points aren't very valuable if you want to do anything but take train trips, but if you do want to take train trips, they're quite valuable. Earning 2.9% in value on unbonused spend with the Starwood Preferred Guest credit card on unbonused spend puts it solidly up there with the most lucrative cashback credit cards.

If you have an upcoming trip you might consider buying points

I earn virtually all of my miles and points through manufactured spend, but I'm perfectly aware that periodic opportunities to pay cash for miles and points can offers discounts off cash rates under certain circumstances. As Frequent Miler explains, there's currently a Starwood Preferred Guest promotion to purchase points for 2.275 cents each, which would offer a discount of 21.6% off the long-haul Amtrak Guest Rewards redemptions I mentioned above.

That's not a huge discount in absolute travel hacking terms, but if you have an upcoming Amtrak trip planned and you'd otherwise pay cash for it, it would be silly not to instead pay 21.6% less for the same trip.

Speculatively transfer points skeptically

I like trains, so I'd happily transfer an almost unlimited number of Starpoints to Amtrak if I were certain they would maintain their current redemption system. Unfortunately, I'm certain they won't, and I wouldn't recommend anyone transfer Starpoints to Amtrak Guest Rewards for train trips they plan to take more than one or two years in the future.

Conclusion

The Starwood-Marriott merger has created a lot of one-time opportunities we'll all be talking about leading up to and after August 1, 2018. The opportunity to transfer Starpoints to Amtrak Guest Rewards is one that anyone with a large Starpoint balance and an interest in Amtrak travel should consider.

Absolute versus relative redemption values

In Wednesday's post about passing on the current IHG Rewards point sale, I mentioned that it's not enough to get good absolute value from a points redemption if you're not also getting good value relative to other redemption options. I think a lot of people understand this idea intuitively, but since it's central to my travel hacking practice, I want to spell it out in more detail.

Absolute value matters if you don't have choices

If you're constrained in your choice of hotels, airlines, or routes, then you are perfectly justified in thinking about the absolute redemption value of your points. A classic example would be a wedding or conference where you're expected to stay in a particular hotel. If the conference rate is $300 per night, and you're able to book it for 30,000 points instead, you know for a fact you're getting 1 cent per point in value.

You still have to make a choice though: is 1 cent per point a good redemption value or a bad redemption value? If you're redeeming a currency that's otherwise redeemable for cash at one cent each, like Ultimate Rewards points, then it's a bad value, since the paid rate will earn a larger rebate than the redemption. If you're redeeming a currency you paid much less than one per point for, then it might be a good value, since you're realizing a discount off a stay you'd otherwise have to pay cash for.

Relative value matters if you get to choose

In Madison, Wisconsin, there are there chain hotels more or less equidistant from the Capitol:

  • Hilton Madison Monona Terrace
  • Hyatt Place Madison/Downtown
  • AC Hotel Madison Downtown

On a random upcoming Wednesday night, the lowest available rates are quite close:

  • Hilton: $144.53
  • Hyatt: $148.01
  • AC Hotel: $162.86

If you were paying cash, you'd book the Hilton and call it a life. Meanwhile, the cost in points is all over the place (as you'd expect since they're different currencies). Here are those costs, and the redemption value compared to our fallback option of paying $144.53 at the Hilton:

  • Hilton: 36,000, 0.4 cents per point
  • Hyatt: 8,000, 1.8 cents per point
  • AC Hotel (Marriott): 25,000, 0.58 cents per point

These are all well within the normal range of redemption values for these currencies. But in order to determine the highest relative value, we need another piece of information: the cost we paid for the currency in question.

If you earn Hilton Honors points through grocery store manufactured spend, you earn 6 points per dollar spent, instead of 2 US Bank Flexpoints (worth 3 cents towards travel redemptions) with the Flexperks Travel Rewards card or 2 Membership Rewards points with the American Express Premier Rewards Gold card. Meanwhile, if your best method of earning Hyatt and Marriott points is transfers from a flexible Ultimate Rewards account like the Sapphire Preferred, Sapphire Reserve, Ink Plus or Ink Preferred, then you're effectively paying one cent each for those points — the value of Ultimate Rewards points when redeemed for cash.

The best relative value is therefore the Hyatt redemption: paying the equivalent of $80 for $144.53 in value is better than paying $180 (Hilton) or $250 (Marriott).

Alternative: availability-weighted relative value

The above methodology is appropriate for someone with access to plentiful manufactured spend and plentiful travel, which is sometimes treated as the "default" mode for travel hackers.

But of course that describes relatively few people in the real world. Far more common are business travelers who passively accrue points balances on their employer's dime, and casual travelers who discover they've accidentally accumulated a substantial balance in one or more loyalty accounts.

In those cases, I think the relative value calculations I described are almost useless, and it's better to use what you might call "availability-weighted" relative value: if Marriott Rewards points are the points you happen to have because your workplace has a contract with Marriott, you should redeem them more aggressively than a strict relative value calculation would suggest.

This is equally true of travel hackers who refuse to redeem points for anything less than their "optimal" value. If you have a large Hilton balance and a low Ultimate Rewards balance, it makes perfect sense to make a weak redemption at the Hilton instead of a good redemption at the Hyatt. That's what I mean by "availability-weighting" relative value.

If this is your strategy, remember you'll also want to normalize your balances for your typical redemption size. If you have 300,000 Hilton Honors points, 300,000 Hyatt points, and 300,000 Marriott points, which currency do you have "more" of? The obvious answer is Hyatt, where redemptions top out at 30,000 points, then Marriott (70,000), then Hilton (95,000+). However, those answers might be flipped if you have particular properties, and particular values, you typically redeem each currency at.

Conclusion

There is so much fuzzy thinking about the value of different loyalty currencies that I usually ignore people trying to nail down the precise value of this or that program, although I liked the Wandering Aramean Hotel Hustle "average point values" feature back when it was functional, mainly because it confirmed my intuitions.

Instead, I find it simpler to work forward from cost rather than backward from value. I know how much I pay for the loyalty currencies I earn, so for a given trip, I try to find the redemptions that cost the least in foregone value, while also taking into account which currencies I have the most of and therefore are most in need of redemption.

Just remember: your least valuable point will always be the one you don't redeem.

You probably shouldn't participate in a good IHG points sale

Through Friday, June 15, IHG Rewards Club is offering a 100% bonus on purchased points, meaning you can buy up to 200,000 points for $1,000, or 0.5 cents each, when you buy 52,000 or more points (lower amounts cost more per point).

This is pretty cheap for IHG Rewards Club points

You can always buy unlimited IHG points for between 0.6 and 0.7 cents each year-round with the "points and cash trick:" reserve a room using points and cash, refund the reservation and your cash co-pay will be refunded in points. So a 50,000-point night that you book with 30,000 points and $130 in cash, then refund, will allow you to buy 20,000 points for $130, or 0.65 cents each.

Still, 0.5 cents is 20% less than 0.6 cents, so if you have upcoming travel plans that are good IHG redemptions at 0.6 cents, they'll be even better redemptions at 0.5 cents each.

You can also earn portal cashback at Points.com, which handles these transactions for IHG, reducing your net cost further below 0.5 cents per point.

IHG Rewards points aren't very valuable, but they're more valuable than that

IHG Rewards points are often worth between 0.5 cents and 1 cent each, so purchasing points at 0.5 cents each speculatively could reasonably be expected to translate into a discount of perhaps 50%, after taxes, on a future stay. That's not bad, especially if you aren't able to manufacture Hilton or Hyatt points in sufficient quantities to meet your travel needs.

Of course, there is a maximum points cost of a free night award of 70,000 IHG Rewards points (for now), while there's no limit to the cost of a paid night, so there's no theoretical limit to the value you can get.

IHG award availability stinks

The problem is that at the properties where you could expect to get the maximum value for your IHG points, award availability is extremely limited. Not only are properties like the InterContinental Hotels Bora Bora Resort Thalasso Spa extremely stingy with award space, what space does become available is quickly snatched up due to IHG's willingness to sell points for so much less than their redemption value at those properties.

A huge pool of cheap IHG points chasing scarce award availability makes people even more desperate to book rooms when they do become available, even holding onto reservations and rooms they don't plan to actually use.

I like the example of the InterContinental in Bora Bora because it has an entire FlyerTalk thread dedicated to folks trying to find award availability there, but the same is true of many desirable IHG properties during periods of even medium demand.

Conclusion

If you have firm plans, have identified an IHG property that provides good absolute value (I'd aim for 1+ cent per point) and relative value compared to any other hotel currencies you have available, and have confirmed the property has reward nights available for the dates you're interested in, then this is an opportunity to buy points at a decent discount compared to the price they're sold for year-round.

But if any of those conditions isn't met, then under no circumstances would I buy IHG Rewards points speculatively.

To my reader who mistyped his own e-mail address, Starwood edition

I recently received a question from a reader using the "Contact" page, but the reader seems to have mistyped their e-mail address, which I discovered only after writing my reply. Hopefully reader TF will stop by and find my answer here instead:

"If you have multiple millions of SPG points and tons of airline miles already then it’s not obvious that you should do anything.

"If there’s a particular airline currency you redeem more than any other, you might consider transferring your SPG points to Marriott and booking a bunch of Hotel + Air redemptions (90k SPG -> 132k United miles or 120k Alaska/Delta/American miles). It’s not clear how Amtrak transfers will work after August 1, so if you like taking Amtrak you can also get a good value transferring your SPG points to Amtrak (note that you don’t get the 5k bonus when transferring 20k points to Amtrak).

"However, it’s not like your points will vanish on August 1. If you have been earning your SPG points through hotel stays, well, you can redeem your new Marriott points for hotel stays going forward. A million SPG points will become 3 million Marriott points, which will be enough for 50 nights at any top-tier Marriott property in the world after August 1 (and a whole lot more nights than that at lower-tier properties).

"If you truly cannot imagine needing any more airline miles, and you truly can’t imagine needing any more Marriott/SPG hotel stays, then you could try to sell points or reservations to other people, or through a points broker. Even if you don’t need a stay or flights, perhaps you have friends or family members you could sell redemptions to at a discount, or be the cool aunt or uncle and send your nieces and nephews on an exotic vacation?

"Let me know if I can help with anything else.

"—FQF"

So, there you go TF.

I try to respond to everyone who leaves comments, e-mails me, or submits a contact form, but if you would like to ask a question in private and not have my answer blasted on the internet, be sure to use a working e-mail address.

Using Chase Ink Business Unlimited to rejigger your Ultimate Rewards card portfolio

Chase has been fiddling with both their personal and business credit card lineup for a few years now, introducing both the personal Freedom Unlimited credit card, which earns 1.5 fixed-value Ultimate Rewards points on all purchases, and the small business Ink Business Preferred, which earns 3 flexible Ultimate Rewards points on "travel, shipping purchases, Internet, cable and phone services, and on advertising purchases made with social media sites and search engines."

The recent launch of an Ink small business version of the Freedom Unlimited means there's potentially an opportunity to use an Ink Business Unlimited card to rejigger your current lineup of Ultimate Rewards-earning credit cards.

Signup bonuses

I don't chase signup bonuses, but I know a lot of my readers do, so if you're eligible for an Ink Business Unlimited credit card, you might want to sign up for one just for its 50,000-point signup bonus after spending $3,000 within 3 months. Ultimate Rewards points are probably the most valuable single currency out there right now, so it's always worth at least considering applying for a signup bonus when a new Ultimate Rewards product is launched, as long as you're eligible.

Product changes

However, not everyone is eligible for new Chase credit cards due to the "5/24 rule," but it should still be possible to get an Ink Business Unlimited card by requesting a product change from an existing Ink credit card.

Personally, I like the ability to earn plentiful Ultimate Rewards points buying Visa prepaid debit cards at Staples and (during promotions) Office Depot and OfficeMax, but if you have several Ink Bold, Ink Plus, or Ink Cash cards you're not maxing out at office supply stores, you might consider requesting a product change from one of those products to an Ink Business Unlimited.

Why? Because the 1.5 Ultimate Rewards points per dollar earned is unlimited, so you only need one of either the Freedom Unlimited or Ink Business Unlimited credit cards. If you're not using the bonus categories on an existing Ink card (even a no-annual-fee Ink Cash card), then switching to an Ink Business Unlimited might free up a personal Freedom Unlimited to be product changed to a valuable Freedom card offering 5 Ultimate Rewards points per dollar at quarterly rotating merchants, or even a Sapphire Preferred or Sapphire Reserve card if you're interested in the travel protection, bonus earning, or redemption uplift of those cards and aren't eligible to apply for them directly (due to 5/24 or simply having already earned the signup bonus too recently).

Referral bonuses

While not necessarily a reason to sign up in its own right, it's also true that Ink Plus and Ink Bold cardmembers can currently only refer friends and fellow business owners to the Ink Business Preferred, while I assume Ink Business Unlimited customers will eventually be able to generate links to Ink Business Unlimited applications (if they can't already).

So if you like earning referral bonuses but don't think the Ink Business Preferred is worth referring people to, you might want a card that can generate links to the no-annual-fee Ink Business Unlimited instead.

Conclusion

Many, if not most, long-time travel hackers are so far above Chase's 5/24 restriction that applying for new Chase credit cards isn't an option. But that's no excuse to ignore the launch of new credit cards, since additional cards open up new opportunities for product changes, and under almost no circumstances should you close a Chase credit card, given the difficulty you might have in opening additional ones in the future.

Did the Marriott Hotel + Air arbitrage opportunity just get even better?

All the way back in September, 2016, I wrote that Marriott Hotel + Air packages offered an outlandishly good opportunity to transfer Starpoints to certain airline (but not train!) partners, even if you never intended to redeem the attached 7-night stay certificate.

A curious post by a Starwood employer on FlyerTalk suggests the deal might have gotten even better.

Use Marriott to transfer Starpoints to certain airline partners

Starpoints can be redeemed in multiples of 20,000 points (up to 80,000 points at a time) to receive the following number of points with a number of airline programs (this is a just an illustrative sample):

  • United MileagePlus: 12,500
  • Alaska Mileage Plan: 25,000
  • Delta SkyMiles: 25,000
  • American AAdvantage: 25,000

Because Starpoints can be converted into three Marriott Rewards points each, Marriott Hotel + Air redemptions allow you to transfer Starpoints into the same programs at slightly different ratios. 270,000 Marriott Rewards points (90,000 Starpoints) can be converted into:

  • United MileagePlus: 132,000
  • Alaska Mileage Plan: 120,000
  • Delta SkyMiles: 120,000
  • American AAdvantage: 120,000

When making these redemptions, you'll also receive a 7-night certificate good at a Category 1-5 property. Due to the fifth night free on award stays, that stay is worth between 45,000 (7 nights at a Category 1 property) and 150,000 points (at a Category 5 property) according to the current award chart.

"Outstanding Marriott Travel Packages will be cancelled and converted to equivalent points"

One thing that could happen after August 1 is that Marriott Travel Packages would be mapped to the new categories, with Category 5 certificates mapped to the new Category 4, Ritz-Carlton Tier 5 certificates mapped to the new Category 8, and the other certificates falling somewhere in between.

But according to Starwood's representative on Flyertalk, that's not the case. Instead:

"Floater certificates, including outstanding Marriott Travel Packages, will be cancelled and converted to equivalent points, credited to the member’s account for future redemption."

You might say (or sing, if you're so inclined), "what do you mean by that? That is not an answer."

And you'd be right! We have no idea how many points Marriott thinks are "equivalent" to a 7-night stay certificate. But instead of whining about uncertainty, I want to walk through a simple exercise thinking instead about the distribution of possibilities:

  • Low-value outcome: Category 1-5 certificates are converted to 45,000 points each;
  • Mid-value outcome: Category 1-5 certificates are converted to between 60,000 and 120,000 points each;
  • High-value outcome: Category 1-5 certificates are converted to 150,000 points each.

This exercise illustrates that just being a "pessimist" isn't enough. Even If you think there's an 85% likelihood of the low-value outcome, a 10% chance of a mid-value outcome, and just a 5% chance of the high-value outcome, you should estimate the value of a 7-night certificate to be around 55,000 Marriott Rewards points, which means you're almost (5,000 transferred Ultimate Rewards points away) at another 25,000-mile transfer under the new Marriott Rewards program.

Of course, the more confident of the mid- or high-value outcomes you are, the more such redemptions you should be willing to make before August 1.

But no matter how you assess the distribution of possibilities, if you typically redeem your Starpoints by transferring them to airline partners, you should consider leaning into this one-time opportunity.

Yes, collecting SkyMiles still makes perfectly good sense, you dolt

Just my luck. The very weekend after I wrote that I was cancelling my Delta Platinum Business American Express because I'd given up on American Express's games awarding high spend bonuses on manufactured spend, Thought Leader From Behind Gary Leff posted another one of his moronic screeds against collecting SkyMiles.

Most people don't know what they're doing, and rightly so

The reason "I started this blog for friends and family" became an immediate cliche to describe affiliate bloggers is that it has a hint of truth in it: our friends and family have no clue what travel hacking is about, and don't have any interest in learning.

What it misses is that it is good and right that most people have not the slightest interest in travel hacking, not because it keeps the game alive, but because most people have better things to do.

Among the 95%+ of the population that doesn't — and shouldn't — care about travel hacking, there's nothing wrong with "collecting" SkyMiles because there's nothing wrong with virtually anything you could suggest.

Obviously at the extreme you could suggest a manifestly inappropriate credit card like a American Express Platinum card, or a Citi Prestige, or a non-rewards credit card like Chase Slate, but the difference between a civilian using a 2% cash back credit card, or an airline credit card, or a hotel credit card, is simply a rounding error to a normal travel budget, and if they prefer to fly on Delta, or prefer to stay at Marriotts, there's no harm in using a co-branded credit card because there's no benefit to using any other card.

Among people who know what they're doing, people will still make mistakes

That brings us to travel hackers, who presumably are Gary's target audience with his harangue, and there are only three logical categories a person can fall in:

  • they can overvalue SkyMiles;
  • they can properly value them;
  • or they can undervalue them.

In other words, when a person decides where a dollar in annual fees, or a dollar in credit card spend, or a dollar in shopping portal purchases, is going to be most lucrative, they can earn more SkyMiles than would be platonically "correct," the correct number of SkyMiles, or fewer SkyMiles than they really ought to.

Gary's whole rant is focused on the mere existence of the first group, those who overvalue SkyMiles. And the first group no doubt exists! I've probably fallen into this group for the last year or so, finding it harder and harder to redeem SkyMiles for the flights I want, while stubbornly sticking with my Delta American Express card to meet the Medallion Qualification Dollar waiver each year.

The problem for Gary's argument is that the second and third groups also exist, and there is no reason obvious to me that the first group is sufficiently larger than the other two groups to merit special attention.

Someone who puts $50,000 in qualifying spend on a Delta Platinum American Express earns 70,000 SkyMiles. To break even against a 2% cash back card they'd need to get an average of 1.43 cents per SkyMiles (weighted by redemption size). But remember, the floor on the value of SkyMiles is 1 cent each when redeemed against paid Delta fares, so the lefthand tail of the distribution of redemptions isn't at 0.0 cents each; it cuts off abruptly at 1.0 cents.

Meanwhile, it's easy to imagine that someone who is able to flexibly and strategically redeem SkyMiles for long-haul and premium cabin seats may not be earning enough SkyMiles. Just like in the case of the over-valuers, a travel hacker may be stuck in the rut of a single Delta Platinum card while they really should be earning many times more SkyMiles using a full suite of personal and business Platinum and Reserve cards, possibly even doubling up with a partner to gift themselves even more Reserve miles.

The existence of errors isn't informative unless you know the distribution of the errors

The reason I don't give advice is that I don't know your circumstances. I think Membership Rewards points, companion tickets, and annual free hotel nights are worth much less than face value, but whether they're worthwhile for you depends on your specific circumstances, about which I have no insight.

I believe many people make the mistake of collecting these instruments because they overvalue them, based on many interactions with my friends and readers.

But I also believe many people properly value them, and I believe many people undervalue them, and don't collect enough of them.

How is this miracle possible? How can I say in the same breath that some people overvalue companion tickets and other people undervalue them? Because they are worth different amounts to different people, and different people value them differently!

A currency is only overvalued if someone puts more value on that currency (sacrifices more value from alternatives) than it is worth to that person. And to know that, you need to know both what they are sacrificing (what their next best alternative is), which we can mostly approximate using widely-available 2%/2.5%/3% cashback credit cards, and how much the miles they earn are "really" worth, which depends on their particularized travel and booking patterns.

Conclusion

I'm the last person I expected to come to the defense of SkyMiles, and hopefully it's obvious I've more or less given up on the program for my own needs. But let's dispel with this fiction that Gary Leff has any insight into the value SkyMiles have in meeting your specific travel needs. If anybody tries to tell you the right program to meet your travel needs without knowing anything about you, it's a sure bet they're looking out for somebody besides you.

Confirmed? American Express is killing the golden Delta goose

I wrote back in February that I was concerned the unbonused manufactured spend I put on my Delta Platinum Business credit card wasn't counting towards my annual Medallion Qualification Dollar waiver, but that since it still appeared properly on my statement I hoped it would at least still earn me a "Miles Boost" of 10,000 Medallion Qualification Miles and 10,000 SkyMiles.

It didn't.

Avoid Simon Malls if you're chasing Amex high spend bonuses

I manufacture most of my unbonused spend at a local Simon Malls location, and obviously American Express has flagged Simon Malls as a merchant where people were buying PIN-enabled prepaid Visa debit cards.

While Simon Malls aren't a bonused merchant with any cards I know of, the activation costs are low enough that it's potentially worth manufacturing spend there with cards that offer a decent return on all spend.

For me, that included the Delta Platinum American Express card, not because 1.4 Delta SkyMiles were worth more than 1.5 Ultimate Rewards points, or 2 cents, but because the Medallion Qualification Dollar waiver and bonus Medallion Qualification Miles made it close enough that I was willing to play along.

While my grocery store spend is still working to trigger the high spend bonuses on my Hilton Honors Ascend card, and so I presume would also trigger Miles Boost with the Delta card, that's a totally unreasonable tradeoff since the spend on the first is bonused and on the second unbonused.

An unreliable card is worse than a bad card

There are still options for manufacturing unbonused spend that may still work to trigger the high spend requirements on American Express cards, like GiftCardMall.com and GiftCards.com, but the problem is that once a company starts targeting obviously manufactured spend, there's no telling where they'll stop.

I was happy to take this chance for the edification of my readers, but you can't build a strategy around praying that a merchant will keep working long enough for you to meet a spend threshold. If you put $25,000 on a Delta Reserve card before the axe falls on your preferred merchant, there's no partial credit; you've wasted the entire $25,000 in spend that could have been put on a more lucrative card.

If reselling is the future, the future is getting closer

I'm not a reseller, except for the occasional stunt or proof of concept, but that's not because I don't think reselling is an important or useful strategy. It's just a strategy for rich folks with garages, and I'm poor and live in an apartment, so have neither the float nor the storage capacity to process the kinds of volume heavy hitters deal in.

Besides its favorable cost structure, the other obvious advantage of reselling in this context is that a reseller's purchases are of actual merchandise from real merchants. That makes unbonused, high-spend opportunities like the Delta American Express cards easy to meet with purchases that have no chance of being flagged as "cash equivalents" or the like.

Conclusion

I find American Express's situation peculiar.

First off, cardholders pay an annual fee, $195 in the case of the Delta Platinum card or $450 for the Delta Reserve. Then merchants pay an additional fee based on the transactions made with the card. Then if the customer doesn't pay their bill off in full each month, they also pay additional interest charges on their balance.

Meanwhile, American Express pays some amount for the SkyMiles they tell Delta to award to customers, plus perhaps an additional bounty when customers trigger their annual high spend thresholds. And, of course, American Express bears the risk of customer defaults on outstanding balances.

It seems to me there are three logical possibilities: either American Express is so bad at negotiating that they are paying more for the SkyMiles awarded on spend than they are earning on that same spend, or they think manufactured spend materially increases the likelihood of default, or both.

Then, of course, there's the illogical possibility: that American Express is making plenty of money off manufactured spend but is cutting off its nose to spite its face anyway.