#73 –
Tuesday, October 29, 2019
http://dennyhatch.blogspot.com/2019/10/73-political-polls-v-direct-mail-tests.html
http://dennyhatch.blogspot.com/2019/10/73-political-polls-v-direct-mail-tests.html
Posted by
Denny Hatch
Political Polls v.
Direct Mail Tests:
Utter Drivel v. Absolute Precision
Utter Drivel v. Absolute Precision
Part
I of our “How the Press Failed You” Series…
“The entire 2016 campaign season was been
characterized by a series of spectacular Silver blunders. Not only did he
notoriously give Hillary Clinton a greater than 99% chance of winning the Michigan primary
(she lost), and bungled Indiana as well, but
he spent much of the past 18 months emitting a series of embarrassing
declarations as well as ludicrous prophecies that totally failed to
materialize. —Nathan J. Robinson, Current Affairs, December 29, 2016
On Monday, November 7, 2016,
polling wunderkind Nate Silver, age 38, was in orbit as the modern era’s most
spectacular predictor of political win and losses.
On the following day, Tuesday, November 8th,
against all odds and predictions, Donald Trump was crowned President of the
United States.
The voters shot Nate Silver out of orbit
and he became a meteorite crashing into Earth creating a mile-wide crater.
In short, overnight Silver went from champ to chump.
My
Opinion: Nate Silver and Vladimir Putin’s
Meddling Cost Hillary Clinton the Presidency
For me, Hillary Clinton was not a
likeable candidate. She came
across as a smug, strident elitist who didn’t bother to visit key states. Because Nate Silver gave
Hillary a 99% chance of winning, it was an excuse for many voters—who could
not stand either candidate—to simply stay home. It was Putin who iced Trump’s crumby,
crummy cake.Meddling Cost Hillary Clinton the Presidency
What Political Pollsters Can Learn from Direct Marketers
Political polling and direct mail testing are basically one in the same.
You contact a small representative
sample of like-minded people on a big list.
From the responses, presumably
you can project the winner of an election—or the number of orders you will
receive on the roll-out of a direct marketing offer.
But, there’s a huge difference.
The
Precision of Direct Mail Testing
Eric
Utne
The year was 1984. East West
Journal editor Eric Utne had an idea for a new magazine. He had two
choices.
1. He could
spend a fortune—$1,000,000.00 or more—to rent office space, hire staff, start
publishing UTNE READER and try to
sell issues by mail and on newsstands.
2. Spend
$150,000 to hire the brilliant copywriting/design team of Bill Jayme and Heikki
Ratalahti to create a direct mail offer and send it to 5,000 names on 20
different carefully chosen lists of known readers of magazines and books.
The
Dry Test
Utne opted for a “dry test”—a powerful FREE offer with a strong letter from the editor for a
product or service that does not exist. It’s a lot cheaper than blowing a
million bucks producing an actual magazine that might bore the hell out of
readers.
Elements of Utne's dry test mailing included:
a brochure describing the features and full-color cover of the proposed premier
issue, a postage-paid Business Reply Envelope, and a personal letter—yet
another Jayme masterpiece.
Bill
Jayme, Legendary, Direct
Mail Copywriter
Here’s
Jayme’s delicious lede:
Jayme’s
Outside Envelope and Order Card
Note the Simple Binary Offer on the Above
Order Card
Choice #1: PLEASE SEND MY FREE COPY and reserve my
one-year subscription.
Choice #2: Trash the mailing.
Unlike the response to a political pollster’s
question, the UTNE READER customer’s positive
response was cast in concrete.
You can
take it to the bank, unlike a political survey where the responder can have
second thoughts in the polling booth after a debate fiasco or an email scandal.
Okay, with
UTNE READER the publisher had no way
of knowing the conversion rate—the number readers who would become paid
subscribers vs. those who would find the final product to be tedious and fail
to pay the invoice.
But at
least Utne had a private universe of folks who raised their hands and
said they wanted to see more.
Two Wildly Different Political Polls
Released on MSNBC the Same Morning
From The New
York Times:
Joe
Biden or Elizabeth Warren? New
Polls Differ
on Who’s Leading 2020 Race
Surveys released this week show varying, and sometimes conflicting,
races unfolding nationally and in early primary states.
A pair of new national polls
present starkly different results. A CNN survey released Wednesday had former Vice
President Joseph R. Biden Jr. with a commanding lead of 15 percentage
points ahead of Senator Elizabeth Warren of Massachusetts. But a Quinnipiac University poll released Thursday had Ms. Warren as the
front-runner, seven points ahead of Mr. Biden.
A number of things can affect a poll’s results, including the
wording of a horse-race question and the order that items are asked. In CNN and
Fox News polls this year, respondents have typically been asked for their
opinions on each of the Democratic presidential
candidates, among other questions, before being queried about their vote preference.
In Quinnipiac’s polls — as well as those conducted by
Monmouth University, which have also shown Ms. Warren climbing steadily —
people have not typically been asked to evaluate the candidates one-by-one
before giving their vote choice.
If such small differences in survey structure are indeed having an effect on results, it may reflect the fact that many respondents are not yet certain about their feelings.
If such small differences in survey structure are indeed having an effect on results, it may reflect the fact that many respondents are not yet certain about their feelings.
—Matt Stevens & Giovanni
Russonello, The
New York Times, 10/24/2019
It’s
the Numbers, Stupid!
According
to Rasmussen
Reports, the total number of registered Democrats as of July 2018 across
all 50 states is 44,242,975.
That’s 44.2 million people.
Now look at the small print at the bottom
left of each of the two charts above:
Quinnipiac: Among 713
Dem Reg. Voters, Oct.
17-21 +/- 4.6 Pts.
CNN: Among 424 Dem Reg. Voters, Oct. 17-20 +/- 5.8 Pts.
These two pollsters queried a total of 1,137 Registered Democratic voters
nationwide.
That’s a micro-minuscule 0.00003528
percent of the Democratic electorate.
Put another way, if the above pollsters
surveyed all 50 states, it means CNN talked to an average of 9 Democratic voters
in each state, while Quinnipiac talked to 14 Democratic voters in each state.
With these teeny-tiny samples
of roiled Democrats involved in the most highly charged and contentious
election in a generation, it’s no wonder the results are not only
contradictory, but also preposterous.
Clearly these were quickie-polls
done on the cheap by organizations desperate for media coverage and
recognition.
CNN should know better.
IMHO, Quinnipiac is a minor little University
down the street from Yale—and pathetically ranked (in a tie) for #153 by US News
& World Report. It has somehow achieved recognition for its PPP
(Political Polling Prowess) and releases the results of its QQQ’s (Queer
Quinnipiac Questionnaires) to the gullible, content-horny media for P.R., Development
and Student Acquisition purposes. So they blew this one big time. Memories are short
and these sad sacks will be b-a-a-c-k next week.
The Idiocy of
MSNBC
In their sick need to make news, Morning Joe released these two absurd
polls side-by-side on the same morning.
NBC’s resident expert geek
Steve Kornacki—who delivers his findings with all the jumping around and manic gesturing
of Huey Long or Beto O’Rourke—said, in effect, “In this case we’re in the
business of choosing polls rather than candidate.”
MSNBC/NBC News and Morning
Joe should be ashamed of themselves for perpetrating this obvious insanity.
Takeaways to
Consider
Attention Direct Markers:
• With an 800-year
history of trial and error, direct marketers have refined the business of
testing down to a gnat’s eyebrow.
• Political pollsters are still in the dark ages.
• Political pollsters are still in the dark ages.
• Why is the basic direct mail test 5000 names? The ballpark
response to a direct mail effort is an average of 2%. Two percent of 5,000 is 100 responses. A list of 100 names is the minimum for statistical accuracy in measuring back-end
results.
• Seattle Direct Marketing Wizard Bob Hacker has come up with a system of
multi-variable
testing that enables marketers to get down-‘n’-dirty quick results with
mailings of far less than 5,000 pieces. Warning: you need a Bob Hacker working
for you to pull it off.
• Beware of greedy list owners who gleefully screw direct marketers.
If you do business with the owner of 1 million names, remember this:
list rental income is free money. It costs nothing to supply a tape of 5,000
names at, say, an average $125/M or a free $625. Turn your entire list of 1
million names 20 times at $125/M, that’s a yummy $2.5 million revenue. That’s supplemental
income—a.k.a. free money.
List owners segregate their
lists using the R-F-M (Recency-Frequency-Monetary Value) formula. The very best names
on a list are those that have bought most Recently and most Frequently and who
spend the most Money. These top tier names are the ones to be coddled, promoted
and loved to death. They can represent 60% of a marketer’s revenue and 80% of the
profits.
If a neophyte marketer orders 5,000
names to test a list, an unscrupulous list owner might supply 5,000 of his top-tier
R-F-M names.
The result: the newbie entrepreneur
will be dazzled by the high response and will order 500,000 of that list on the roll-out.
The list owner will supply 500,000 of his least profitable names that will
generate vastly poorer results.
The list owner wins big; the entrepreneur is screwed.
The list owner wins big; the entrepreneur is screwed.
• Any inexperienced entrepreneur (e.g. a retailer) getting into the direct
marking arena is nuts not to hire a world-class list expert as well as top
creative people.
Attention Political Reporters (and Candidates):
• When seeing new polling results in print, online or on TV, the first
thing you should look for is the number of respondents.
• Only once before in recent American history has a dark- dark-dark horse
presidential candidate defeated an odds-on favorite.
The election was 1944 when President Harry S Truman was so far down in the polls against New York Governor Thomas E. Dewey that the Gallup Organization flat-out ceased covering him.
The election was 1944 when President Harry S Truman was so far down in the polls against New York Governor Thomas E. Dewey that the Gallup Organization flat-out ceased covering him.
November 3, 1948. Under tight deadline, The Chicago Tribune was suckered into believing the projections and went to press early, only to be humiliated by this bogus headline.
If you are thinking of running for office—or are
working on a political campaign—for Congress, governor or local office—Philip
White’s WHISTLE STOP provides a wiring diagram for victory. What worked
in 1948 will work like gangbusters today, whether your candidate is running
locally, statewide or nationally.
WHISTLE STOP is delicious reading—a rip-snorting
page turner you won’t be able to put down. Guaranteed!
###
Word count:1742
