Some online experts advise the use of synthesised shielded twin instead of ordinary two wire line for HF antennas claiming it is vastly superior.
Now it could be vastly superior
for several reasons in all, but let's focus on just one important parameter, loss under mismatch conditions.
The scenario then is the very popular 132′ multi band dipole:
 the famous 40m (132′) centre fed dipole;
 20m of feed line being parallel RG6/U CCS quad shield with shields bonded at both ends;
 7MHz where we will assume dipole feed point impedance is ~4000+j0Ω.
We will consider the system balanced and only deal with differential currents.
Now rather than depend on loss calculators, most of which don't reconcile with measurement of CCS RG6/U, I will used measured loss. RG6/U with CCS centre conductor at HF gives a chart of measured loss of a sample of commercial grade CCS quad shield coax.
Above is a comparison of matched line loss (MLL) based on measurement of a length of RG6/U Quad Shield CCS cable and prediction from Simsmith of Belden 8215 (also CCS). The ripple is due to measurement system error, measurements were made quite some years ago with a AIMuhf.
Scaling from the chart, MLL @ 7MHz is 0.039dB/m.
Now the MLL of the 150Ω shielded twin formed by the parallel RG6/U coaxes is exactly the same as the component coaxes, MLL @ 7MHz is 0.039dB/m, and half the load impedance is seen by each coax.
Next step is to calculate the loss under mismatch. For that we will use RF Arbitrary Transmission Line Loss Calculator and specify the conductor loss as 0.039dB/m (at 7MHz, it is reasonable to allocate all of the MLL to conductor loss).
Above is the calculator input form.
Parameters  
Ro  75.00000 
vf  0.800 
Conductor loss (dB/m)  3.900000e2 
Dielectric loss (dB/m)  0.000000e+0 
k1  1.463644e5 
k2  0.000000e+0 
Frequency  7.100 MHz 
Length  20.0000 m 
Zload  1.000e+3+j0.000e+0 Ω 
Yload  1.000e3+j0.000e+0 S 
Results  
Zo  75.02j1.81 Ω 
Length  213.210 °, 0.592249 λ, 20.000000 m, 8.342e+4 ps 
Line Loss (matched)  0.780 dB 
Line Loss  3.178 dB 
Efficiency  48.11 % 
Zin  3.614e+1j1.063e+2 Ω 
Yin  0.002867+j0.008433 S 
VSWR(50)in  8.24 
That reports 3.178dB of loss under the specified mismatch conditions. Note that this RG6/U is the pretty common CCS product used for TV and satellite TV installations, you can buy poorer cable on eBay (RG6 cladding thickness report).
If the load were 4000+j0Ω, the loss under mismatch would be 4.955dB.
Lets evaluate the loss using a home made open wire line of 2mm copper conductors spaced 150mm. This time of course, the line sees all of the 2000+j0Ω load.
Above is the calculator input form.
20m of open wire line 

Parameters  
Conductivity  5.800e+7 S/m 
Rel permeability  1.000 
Diameter  0.002000 m 
Spacing  0.150000 m 
Velocity factor  1.000 
Loss tangent  0.000e+0 
Frequency  7.000 MHz 
Twist rate  0 t/m 
Length  20.000 m 
Zload  2.000e+3+j0.000e+0 Ω 
Yload  5.000e4+j0.000e+0 S 
Results  
Zo  602.02j0.76 Ω 
Velocity Factor  1.0000 
Length  168.116 °, 0.466990 λ, 20.000000 m, 6.671e+4 ps 
Line Loss (matched)  3.21e2 dB 
Line Loss  6.11e2 dB 
Efficiency  98.60 % 
Zin  1.411e+3+j8.329e+2 Ω 
Yin  5.256e4j3.102e4 S 
That reports 0.061dB of loss under the specified mismatch conditions, less than 2% of the RG6/U twin scenario.
If the load were 4000+j0Ω, the loss under mismatch would be 0.12dB.
With such a significant difference in line loss under the mismatch scenario, it is hard to understand why anyone might reasonably claim the RG6/U twin is vastly superior.
An exercise for the reader, rework the shielded twin using your choice of better coax. Is it good enough?